Digital Contact Tracing for Pandemic Response Kahn Jeffrey Johns Hopkins Project on Ethics and Governance of Digital ContactTracing Technologies
Published by Johns Hopkins University Press
Kahn Jeffrey and Johns Hopkins Project on Ethics and Governance of Digital Contact Tracing Technologies Digital Contact Tracing for Pandemic Response Ethics and Governance GuidanceJohns Hopkins University Press 2020 Project MUSE doi101353book75831 httpsmusejhuedu
For additional information about this book
[ Access provided at 8 Oct 2021 1139 GMT with no institutional affiliation ]
This work is licensed under a
httpsmusejhuedubook75831
Creative Commons Attribution 40 International License
DIGITAL CONTACT TRACING FOR PANDEMIC RESPONSE
DIGITAL CONTACT TRACING FOR PANDEMIC RESPONSEEthics and Governance Guidance
Johns Hopkins University PressBaltimore
Edited by Jeffrey P Kahn PhD MPH Chair
Johns Hopkins Project on Ethics and Governance of Digital Contact Tracing Technologies
copy 2020 Johns Hopkins University Press
This work is also available in an Open Access edition which is licensed
under a Creative Commons AttributionndashNonCommercialndashNoDerivatives 40
International License httpscreativecommonsorglicensesby-nc-nd40
All rights reserved Published 2020
Printed in the United States of America on acid-free paper
9 8 7 6 5 4 3 2 1
Johns Hopkins University Press
2715 North Charles Street
Baltimore Maryland 21218-4363
wwwpressjhuedu
Library of Congress Cataloging-in-Publication Data is available
ISBN 978-1-4214-4061-3 (paperback acid-free paper)
ISBN 978-1-4214-4062-0 (ebook)
ISBN 978-1-4214-4063-7 (ebook open access)
Special discounts are available for bulk purchases of this book For more information
please contact Special Sales at specialsalespressjhuedu
Johns Hopkins University Press uses environmentally friendly book materials
including recycled text paper that is composed of at least 30 percent post-
consumer waste whenever possible
Contents
Lead Authors and Contributors vii
Preface ix
Acknowledgments xi
Acronyms and Abbreviations xiii
Summary 1
Introduction 1
DCTT Features Functions and Potential Applications 3
Summary of Recommendations 9
Summary of Analysis 13
Introduction 23
Guiding Principles for the Use of Digital Public Health
Technologies for Pandemic Response 25
1 Public Health Perspective 29
Types of Information Collected through Contact Tracing 29
How Contact Tracing Information Informs Public Health Action 31
Characteristics That Make Data Useful to
Public Health for Reducing Disease Transmission 32
2 Digital Technology and Contact Tracing 35
Characteristics of SARS-CoV-2 Relevant to Candidate Digital Solutions 35
Previously Existing Contact Tracing Technologies 36
Introduction of Novel Digital Contact Tracing Technologies 37
Relevant Differences between Manual and Digital Contact Tracing 41
vi Contents
3 Ethics of Designing and Using DCTT 43
Justifying the Use of DCTT Systems 44
Monitoring and Evaluating Technologies to Inform Policy and Practice 45
Public Trust and Public Attitudes 48
Designing Flexible Technology to Maximize Public Health Utility
While Respecting Other Values 50
Policy Positions to Advance Widespread Use of Digital Contact
Tracing Technologies 59
Disclosure and AuthorizationConsent 63
Promoting Equity and Fairness 69
Instituting Transparent Governance and Oversight 72
4 Legal Considerations 75
Data Privacy and Data Security Laws 77
Health Information Privacy 82
Labor and Employment Privacy Rights 84
Constitutional Privacy Rights 86
Consent 93
Anti-discrimination and Individual Freedom Laws 94
5 Recommendations 97
Public Health 97
Ethics 98
Legislative 102
Resources 103
US Government Response 103
Other Governmental and Nongovernmental Organizations 106
Digital Contact Tracing Experiences from Other Countries 112
Specific Digital ProductsApps 115
Polling 117
Popular Press 119
Commentaries 120
Academic Literature 121
Works Cited 127
Lead Authors and Contributors
vii
Lead Authors
Joseph Ali JD Assistant Professor Department of International Health Johns Hopkins Bloomberg School of Public Health Core Faculty amp Associate Director for Global Programs Johns Hopkins Berman Institute of Bioethics
Anne Barnhill PhD Core Faculty amp Research Scholar Johns Hopkins Berman Institute of Bioethics
Anita Cicero JD Deputy Director Johns Hopkins Center for Health Security Visiting Faculty Johns Hopkins Bloomberg School of Public Health
Katelyn Esmonde PhD Hecht-Levi Postdoctoral Fellow Johns Hopkins Berman Institute of Bioethics
Amelia Hood MA Research Program Coordinator Johns Hopkins Berman Insti-tute of Bioethics
Brian Hutler PhD JD Hecht-Levi Postdoctoral Fellow Johns Hopkins Berman Institute of Bioethics
Jeffrey Kahn PhD MPH Andreas C Dracopoulos Director Johns Hopkins Ber-man Institute of Bioethics
Alan Regenberg MBE Director of Outreach amp Research Support Associate Fac-ulty Johns Hopkins Berman Institute of Bioethics
Crystal Watson DrPH MPH Senior Scholar Johns Hopkins Center for Health Security Assistant Professor Department of Environmental Health and Engi-neering Johns Hopkins Bloomberg School of Public Health
Matthew Watson Senior Analyst Johns Hopkins Center for Health Security Senior Research Associate Department of Environmental Health and Engi-neering Johns Hopkins Bloomberg School of Public Health
Other Contributors
Robert Califf MD MACC Head of Clinical Policy and Strategy Verily and Goo-gle Health
Ruth Faden PhD MPH Philip Franklin Wagley Professor of Biomedical Ethics amp Founder Johns Hopkins Berman Institute of Bioethics
viii Lead Authors and Contributors
Divya Hosangadi MSPH Senior Analyst Johns Hopkins Center for Health Secu-rity Research Associate Department of Environmental Health and Engi-neering Johns Hopkins Bloomberg School of Public Health
Nancy Kass ScD Deputy Director for Public Health amp Phoebe R Berman Pro-fessor of Bioethics and Public Health Johns Hopkins Berman Institute of Bioethics
Alain Labrique PhD MHS MS Professor Department of International Health Johns Hopkins Bloomberg School of Public Health Director Johns Hopkins Global Health Initiative
Deven McGraw JD MPH LLM Chief Regulatory Officer CiitizenMichelle Mello JD PhD Professor of Law Stanford Law School Professor of
Health Research and Policy Stanford University School of MedicineMichael Parker BEd (Hons) MA PhD Director Wellcome Centre for Ethics and
Humanities Ethox Centre University of OxfordStephen Ruckman JD MSc MAR Senior Advisor to the President for Policy
Office of the President Johns Hopkins UniversityLainie Rutkow JD MPH PhD Senior Advisor to the President for National Cap-
ital Academic Strategy Office of the President Johns Hopkins UniversityJosh Sharfstein MD Vice Dean for Public Health Practice and Community
Engagement Professor of the Practice Johns Hopkins Bloomberg School of Public Health
Jeremy Sugarman MD MPH MA Deputy Director for Medicine Harvey M Meyerhoff Professor of Bioethics and Medicine Johns Hopkins Berman Institute of Bioethics Department of Medicine Johns Hopkins School of Medicine and Department of Health Policy and Management Johns Hop-kins Bloomberg School of Public Health
Eric Toner MD Senior Scholar Johns Hopkins Center for Health Security Senior Scientist Department of Environmental Health and Engineering Johns Hop-kins Bloomberg School of Public Health
Marc Trotochaud MSPH Analyst Johns Hopkins Center for Health Security Research Associate Department of Environmental Health and Engineering Johns Hopkins Bloomberg School of Public Health
Effy Vayena PhD Professor Health Ethics amp Policy Lab Department of Health Sciences amp Technology ETH Zurich
Tal Zarsky JSD LLM LLB Professor of Law University of Haifa Faculty of Law Visiting Scholar University of Pennsylvania Law School (2019ndash2020)
Preface
ix
Digital technologies are being developed and promoted to support the public health response to the COVID-19 pandemic with discussion and implementation planning in the United States by localities states institu-tions and employers Key decision makers and stakeholdersmdashincluding government officials institutional leaders employers digital technology developers and the publicmdashrequire clear and well-supported guidance to inform the deployment and use of these technologies as well as of the data they collect store and share While technology-based approaches are currently unable to provide solutions on their own experiences in other countries indicate that they could be used successfully in conjunc-tion with traditional and novel public health methods
This report reflects a rapid research and expert consensus group ef-fort led by the Berman Institute of Bioethics and the Center for Health Se-curity at Johns Hopkins University It draws on experts from both inside and outside Johns Hopkins in bioethics health security public health technology development engineering public policy and law The report highlights issues that must be addressed and provides recommendations for the use of digital technologies as part of contact tracing
The analysis offered here is focused on answering the following questions
bull Can digital contact tracing technologies (DCTT) be effective as part of public health responses to the pandemic and if so to what degree for which specific types of functions with what confidence and with what requirements
bull How can these technologies serve the interests of public health while respecting other individual and collective interests such as ensuring equitable distribution of benefits and burdens and limit-ing infringement on privacy and other civil liberties
x Preface
bull What are the ethical legal policy and governance guardrails cur-rently in place around such technologies and what else is needed
bull What additional guardrails are required to ensure that the goals of public health in using these technologies are achievable in ways that are ethically and legally sound
To answer these questions the report examines some core aspects of dig-ital technologies applied to contact tracing focusing on
bull the value of and basic methods for traditional public health surveil-lance and contact tracing
bull candidate technological products to enhance public health surveil-lance and contact tracing how they work and their comparative value for public health
bull core ethical legal and governance considerations and how they relate to relevant features of candidate technological solutions and
bull what is needed to move forward responsibly with the use of digital technology in support of public health surveillance acknowledg-ing gaps in our current understanding
The project involved in-depth analysis by a dedicated team of faculty postdoctoral fellows and research staff working over the course of only a few weeks but with great intensity drafting a report in collaboration with 26 total contributors writing commenting and revising through multiple drafts with the penultimate draft ldquopressure-testedrdquo by review and discus-sion at a virtual workshop of invited experts and stakeholders held on May 13 2020 and the final version completed on May 21 2020 The report and analysis builds on the excellent work of others in some parts of this territory while focusing on the gaps in analysis and areas that have not been sufficiently addressed The goal is to offer comprehensive guid-ance to relevant stakeholders to advance public health response during the COVID-19 pandemic Given the rapidly evolving territory into which DCTT is being introduced this report will by necessity be something of a living document updated as often as information dictates in order to continue to offer leading-edge analysis and guidance Versions will be noted in the digital and print editions
Acknowledgments
xi
Efforts like this project require teams and even small armies to be carried out successfully and this was no exception except that it was many fewer people working many more hours than could reasonably be expected of them From the initial kernel of an idea to the publication of this report in book form this project took just over a month total That seems impossi-ble even as I know it is accurate and it speaks to the incredible commit-ment hard work research skills and analytic acumen of our colleagues at Johns Hopkinsmdashthe core team are deservedly listed as lead authors of this report
None of this would have been possible without the supportmdashmoral and financialmdashand encouragement of Johns Hopkins University Presi-dent Ronald J Daniels who was the first to suggest the idea to me of taking on this topic He provided not only support and encouragement but the imprimatur of his office including help guidance and counsel from Prof Lainie Rutkow senior advisor to the president Lainie played a more integral role than that description captures reflected in part by her inclusion among the reportrsquos contributors but she deserves special acknowledgement for shepherding us through to the end
I mentioned that this was a team effort and every team requires an effective leader My colleague Prof Joseph Ali stepped into that role as we undertook the project and then he worked seven days a week along with the rest of the core research and writing team always unfailingly positive and deeply engaged in the work He along with Prof Anne Barnhill Alan Regenberg Amelia Hood and Drs Katelyn Esmonde Brian Hutler and Crystal Watson all deserve special thanks for doing so much in so little time all while working under the grinding social distancing restrictions of the 2020 pandemic That work was supported by Arnold amp Porter Kaye Scholer LLP with legal research and other assistancemdasha huge thanks to
xii Acknowledgments
them Finally the 16 contributing authors were incredibly generous with their time energy and insights all on ridiculously tight timelines and never a complaint or objection
The project benefited greatly from a number of experts who provided written feedback on drafts and who attended the virtual workshop to test our recommendations including Miles Stewart Rob Nichols Smisha Aagarwal Karl Steiner Anupam Joshi Charles Scheeler Ford Rowan and Jay Wagley
Last the fact that this report appears in published book form by Johns Hopkins University Press is another minor miracle from manu-script to printed book in under a week Thanks to JHUP Director Barbara Kline Pope and her team for being willing to take on the challenge and for the incredible focused effort it required
my heartfelt thanks andappreciation to you all
Jeffrey Kahn
Acronyms and Abbreviations
xiii
ADA Americans with Disabilities ActBLE Bluetooth Low EnergyCalOPPA California Online Privacy Protection ActCBP Customs and Border ProtectionCCPA California Consumer Privacy ActCDC Centers for Disease Control and PreventionCLOUD Act Clarifying Lawful Overseas Use of Data ActCOPPA Childrenrsquos Online Privacy Protection ActCOV+ confirmed positive SARS-CoV-2 test resultCOVID-19 coronavirus disease 2019CPNI customer proprietary network informationCSLI cell-site location informationDCTT digital contact tracing technology and closely related
digital health productsECPA Electronic Communications Privacy ActEEOC Equal Employment Opportunity CommissionE-SIGN Electronic Signatures in Global and National
Commerce ActEU European UnionFCC Federal Communications CommissionFTC Federal Trade CommissionFTCA Federal Trade Commission ActGIS geographic information systemGPS global positioning systemHHS US Department of Health and Human ServicesHIPAA Health Insurance Portability and Accountability ActHIV human immunodeficiency virusICU intensive care unit
xiv Acronyms and Abbreviations
IRB institutional review boardJHU Johns Hopkins UniversityOCR Office for Civil Rights US Department of Health amp
Human ServicesOSHA Occupational Safety and Health AdministrationPHI protected health informationPII personally identifiable informationPPE personal protective equipmentPPPT privacy-preserving proximity trackingQR Code quick response codeRFRA Religious Freedom Restoration ActRLUIPA Religious Land Use and Institutionalized Persons ActSARS-CoV-2 severe acute respiratory syndrome coronavirus 2SCA Stored Communications ActSTI sexually transmitted infection
DIGITAL CONTACT TRACING FOR PANDEMIC RESPONSE
11
Summary
Introduction
Public health professionals around the world are working tirelessly to respond to the COVID-19 pandemic using tried-and-tested public health methods for infectious disease surveillance and control These traditional methods are essential to the global COVID-19 response To complement these actions and potentially augment the speed and efficacy of the pub-lic health workforce digital technologies are being harnessed Given the scale of the pandemic significant efforts are being undertaken to develop and leverage public-facing and health-system-supportive technology solu-tions including smartphone apps and other digital tools that may aid public health surveillance and contact tracing
Digital contact tracing technology and closely related digital health products (together DCTT) have been used in several countries as part of broader disease surveillance and containment strategies In the United States DCTT has been proposed as an integral part of some plans to ldquoreopenrdquo the country (Allen et al 2020 Hart et al 2020 Simpson and Conner 2020) It is almost certain that these and related technologies will become part of not only the COVID-19 response but also the larger toolbox for future public health communicable disease prevention and control
These technologies have significant promise They also raise import-ant ethical legal and governance challenges that require comprehensive analysis in order to support decision-making Government officials pub-lic health leaders leaders of institutions employers digital technology de-velopers and the public all must be adequately informed in order to make
2 Digital Contact Tracing for Pandemic Response
responsible choices Johns Hopkins University recognized the importance of helping to guide this process It organized an expert group with mem-bers from inside and outside of Hopkins and led by its Berman Institute of Bioethics in collaboration with the Center for Health Security Its charge was to examine the ethics law policy and public health implications of using digital technologies as part of pandemic response and to develop guidance including a framework and actionable recommendations for governmental and institutional decision makers
Overall this expert group urges a stepwise approach that prioritizes align-ment of technology with public health needs and public values building choice into design architecture and capturing real-world results and impacts to allow adjustments as required Further we urge an approach that recog-nizes that there are complicated issues to resolve for governments insti-tutions and businesses and that introduction of DCTT must include public engagement and ongoing assessments to improve both performance and adoption
Specific recommendations include the following
bull There is no ldquoone size fits allrdquo approach to DCTT Technology design should not be static but it should be capable of evolving depending upon local conditions new evidence and changing preferences and priorities
bull Technology companies alone should not control the terms condi-tions or capabilities of DCTT nor should they presume to know what is acceptable to members of the public
bull DCTT should be designed to have a base set of features that pro-tect privacy with layers of additional capabilities that users may choose to activate An initial default should be that user location data are not shared but users should be provided with easy mech-anisms and prompts to allow for opting-in to this capability with encouragement to the public if it is shown to be critical to achieving public health goals
bull Data collected through DCTT should be made available to public health professionals and to researchers in de-identified form to support population-level epidemiologic analyses
Summary 3
bull Those who authorize use of DCTT within a particular jurisdic-tion or institution should continuously and systematically monitor the technologyrsquos performance in that context This should include monitoring for effectiveness and benefit monitoring for harms and monitoring for the fair distribution of both benefits and harms
bull Governments should not require mandatory use of DCTT given uncertainty about potential burdens and benefits Additional tech-nology user and real-world testing is needed
Through in-depth analysis and recommendations this report seeks to guide decision-making and enhance understanding of
bull the value of and basic methods for traditional public health surveil-lance and contact tracing
bull candidate technological products to enhance public health surveil-lance and contact tracing and their comparative value for public health
bull core ethical legal policy and governance considerations and how they relate to relevant features of candidate technological solutions and
bull what is needed to move forward responsibly with the use of digital technology in support of public health surveillance acknowledg-ing gaps in our current understanding
The full set of recommendations are intended to (1) support effective and informed adoption of DCTT (2) encourage design of flexible technol-ogies that maximize public health utility while respecting other values (3) establish meaningful processes for user disclosure and authorization (consent) (4) promote equity and fairness in the uses of DCTT and (5) foster transparent governance and oversight
DCTT Features Functions and Potential Applications
Digital contact tracing technologies and platforms can be roughly catego-rized into three broad approaches along a spectrum of potential policies and methods a maximal approach (typified by the South Korean govern-
4 Digital Contact Tracing for Pandemic Response
mentrsquos centralized and triangulated data collection (M S Kim 2020)) a minimal approach (typified by the AppleGoogle decentralized priva-cy-preserving proximity tracking (PPPT) and contact notification (Apple and Google nd)) and a diverse range of middle-ground approaches that aim to augment manual contact tracing with the collection of digital data that can be shared with public health authorities
Minimal approaches such as the AppleGoogle PPPT use Bluetooth Low Energy (BLE) ldquohandshakesrdquo that record close contact between mo-bile phone users but do not register the location in which the contact hap-pened In most architectures these proximity data are stored in the usersrsquo phones as anonymized ldquobeaconsrdquo that cannot be used to re-identify the users directly If a user with a PPPT app installed on their phone tests pos-itive and enters test results into their app those who have been in contact with them can be notified by the app This ldquoexposure notificationrdquo can be automatic or at the discretion of the COV+ person depending on the app design If notified a user who has been in contact with a COV+ individ-ual would receive a push notification alerting them to possible exposure (which may be timestamped) but with no other identifying information
The most prevalent middle-ground approach in the US context in-volves the collection and storage of personal datamdashincluding identifying information and location datamdashon the userrsquos phone These decentralized but personally identifiable data can then be voluntarily shared with pub-lic health officials if the user tests positive for SARS-CoV-2 (severe acute respiratory syndrome coronavirus 2) For example a team at the Mas-sachusetts Institute of Technology (MIT) has developed an app called Private Kit Safe Paths (MIT nd) that stores usersrsquo location data on their phone for 28 days If a user tests positive she can voluntarily upload her location data to a website that is accessible only to public health officials Officials can then analyze these personally identifiable data and subse-quently broadcast redacted and de-identified data to other users Healthy users would have access to these redacted location data of COV+ users but their own data would not leave their phones At a minimum the storage of user location data can function as a ldquomemory aidrdquo if the user tests positive but releasing the data to public health authorities may help to analyze the spread of SARS-CoV-2 and alert individuals or groups that have been in contact with COV+ patients
Summary 5
The US Centers for Disease Control and Prevention (CDC) has pub-lished preliminary criteria for evaluating capabilities and attributes of DCTT (CDC 2020e) These and other resources suggest that a compre-hensive assessment of DCTT and its potential to advance the publicrsquos health will require careful consideration of numerous interconnected fac-tors that interact in complex ways and must be navigated within the chal-lenging contexts of uncertainty and urgent need (Figure 1) These include
bull scientific and epidemiological understanding of SARS-CoV-2 transmission and infection
bull public health needs for combating the outbreak
bull technological capabilities of DCTT
bull performance of DCTT applications
bull ethical values and principles
bull characteristics of public adoption and acceptance and
bull legal issues and landscape
FIGURE 1 Interrelating Factors That Frame Responsible Development of Digital Contact
Tracing Technology
6 Digital Contact Tracing for Pandemic Response
The primary objectives for use of DCTT during the COVID-19 pandemic must be to reduce illness and death and facilitate public health efforts to reduce transmission of the virus These objectives fall under a broader overall goal of contributing to societal well-being during the pandemic It is not yet known whether and how much DCTT can contribute to these primary objectives nor whether it will be able to contribute without gen-erating new burdens or even harms such as incorrect warnings or ldquonoiserdquo that detract from the work of manual contact tracing
The process of identifying acceptable technology designs and uses is complex given the interplay among the factors Our analysis reveals that there is no ldquoone size fits allrdquo approach to DCTT There is variability across the United States with respect to SARS-CoV-2 prevalence and in-fection rates public health capacity public attitudes toward DCTT and acceptability of various potential features Moreover our understanding of SARS-CoV-2 and DCTT is evolving public health response needs and capabilities are changing and public attitudes are shifting Different tech-nologies used in different ways may be appropriate to achieve slightly dif-ferent public health goals in different localities and at different points in the pandemic A tiered and phased approach to technology development should be facilitated by law and policy prioritizing underlying interoper-ability while permitting user choices now and for the future
Given the complexity of the terrain as a first step those developing or considering widespread use of DCTT as part of pandemic response should be guided by the following principles and related actions (see box) These principles are meant to apply to DCTT as well as other dig-ital technologies used in novel ways during pandemic response
These principles make clear that in order to maximize the public good from use of DCTT public health needs and technological capabili-ties must be carefully aligned Government officials public health leaders leaders of other institutions employers digital technology developers and the public are all key stakeholders that must be informed and en-gaged in order to enable the most successful and ethically acceptable uses of DCTT
Guiding Principles for the Use of Digital Public Health Technologies for Pandemic Response
Transparency and public engagement are essential to an inclusive digital public health response
bull Government public health and digital technology leaders must engage
effectively with the public and other stakeholders to communicate the util-
ity importance oversight and limitations of relevant digital technologies
including their implications for individualsrsquo privacy and civil liberties
bull Transparency at all levels is essential for maintaining public trust and
confidence
bull To the extent possible digital public health responses should reflect the
range of values that are important to individuals including advancing the
health and well-being of the community as a whole
bull Decision makers should recognize the sacrifices that some people may be
willing to make during a pandemic in order to advance public health goals
Acceptance by some of particular monitoring capabilities should not be read
as a willingness to extend these methods to other problems or uses
Digital public health responses must represent the least infringement of civil liberties necessary to accomplish the public health goals
bull If preferred digital public health strategies infringe on privacy and other civil
liberties the infringements must be sufficiently justified by the circum-
stances of the pandemic offset by ample anticipated public benefit and
considered relative to infringements associated with other possible strate-
gies such as mass physical distancing
bull Only those data that are necessary and relevant for the stated public health
purposes should be collected Identifiable data should be stored in a secure
manner and only for the period of time that the public health purposes
require
bull Adopted technologies should not be used in ways that subject communities
to discrimination or surveillance for nonndashpublic health reasons
bull Respect for individual autonomy requires that users are sufficiently informed
of the public health goals of the technology and the extent to which those
goals are being met
Use of digital public health technologies and data must be guided by best available evidence
bull Decisions to deploy digital public health technologies should be based on a
careful assessment of the uses and limitations of any proposed technology
taking into account the best available evidence
bull Those who deploy digital public health technologies should continuously and
systematically monitor their performance as well as any evidence that is being
generated in other contexts about the selected technological solution and about
other competing technologies
bull Unintended consequencesmdashincluding those that might impact public health
goals core values and interests of the public and unfair advantage or disadvan-
tagemdashshould be carefully monitored and addressed as necessary
Responsible use of digital public health technology requires meaningful governance and accountability
bull Systems of governance must be trustworthy and well informed They must be
reviewed and adjusted as circumstances and evidence change or as unintended
effects are identified
bull Trusted representatives who are capable of developing and implementing uni-
form and fair standards for adopting and utilizing underlying digital technology
must be identified
bull Understandable transparent and publicly accessible rules must guide the
collection access control use storage and combination of data by government
authorities public and private institutions and other parties such as public
health researchers
bull Oversight accountability and consequences for abuse or misuse of these data
must be explicit and enforceable
The deployment of digital public health technology must be rooted in a commitment to equity
bull Digital public health technologies should be deployed in a manner that does
not propagate preexisting patterns of unfair disadvantage or further distribute
harms and risks unfairly throughout the population
bull To the extent possible digital public health technologies should be designed to
rectify existing inequities
bull Oversight mechanisms must be in place to ensure that the improved public
health outcomes are equitable and to detect and correct any unforeseen resul-
tant injustices attributable to the technology or that can be addressed using the
technology
bull The incentives and disincentives for adopting new technology must be equita-
ble not exploitative and aligned with effective use of the technology
bull Disparity-driven technology gaps should be explicitly recognized To the extent
possible provisions should be made to address the digital divide
Summary 9
Summary of Recommendations
The guidance document makes a number of recommendations related to (1) supporting effective and informed adoption of DCTT (2) designing flexible technologies to maximize public health utility while respecting other values (3) establishing meaningful processes for user disclosure and authorizationconsent (4) promoting equity and fairness in application of DCTT and (5) instituting transparent governance and oversight Here we provide a summary of recommendations
Supporting Effective and Informed Adoption
bull Those who authorize use of DCTT within a particular jurisdic-tion or institution should continuously and systematically monitor the technologyrsquos performance in that context This should include monitoring for effectiveness and benefit monitoring for harms and monitoring for the fair distribution of both benefits and harms They should also monitor evidence that is being generated in other contexts about their selected technological solution and about other competing technologies
bull Data collected through DCTT should be made available to public health professionals and to researchers in de-identified form to support population-level epidemiologic analysis
bull Data should be available to users that would permit them to further investigate their personal risk with public health officials or other health workers to add a layer of protection against unnecessary quarantine
bull Technologies or apps may produce some false negatives or false positives but they should be accurate enough that public health authorities feel confident that they support and donrsquot detract from contact tracing efforts
bull Trusted leaders should be enlisted to communicate effectively with the public about DCTT and encourage its use should the technol-ogy demonstrate some potential The limits of knowledge regard-ing effectiveness should also be explained along with what will be done to improve technological capabilities as understanding evolves
10 Digital Contact Tracing for Pandemic Response
bull Incentives can be a useful complement to encouragements how-ever any incentives for users to install and use DCTT must be equitable should not be coercive and should align with effective use of the technology
bull DCTT use should not be mandated at this time given uncertainty about potential harms and benefits Additional technology user and real-world testing is needed
Designing Flexible Technology to Maximize
Public Health Utility While Respecting Other Values
bull Technology companies should not alone control the terms condi-tions or capabilities of DCTT nor should they presume to know what is acceptable to members of the public
bull A ldquovalues in designrdquo approach to development of DCTT should be adopted (Flanagan Howe and Nissenbaum 2008 Knobel and Bowker 2011) Robust public and user engagement activities should be pursued to identify and incorporate to the extent pos-sible a range of values into the design of the technology These values may include privacy but also autonomy efficiency equity or others Technology design should reflect an appropriate balance and prioritization of identified values
bull Technology design should not be static but should be capable of evolving depending upon local conditions new evidence and changing preferences and priorities
bull DCTT should be designed to have a base set of features that pro-tect privacy with layers of additional capabilities that users may choose to activate An initial default should be that user location data are not shared but users should be provided with easy mech-anisms and prompts to allow for opting-in to this capability with encouragement to the public if it is shown to be critical to achieving public health goals
Establishing Meaningful Processes for
User Disclosure and Authorization (Consent)
bull A clear and concise module consisting of basic disclosure and vol-untary authorization should be developed to accompany DCTT
Summary 11
This module should not take the form of ldquoclickwraprdquo terms of service or end-user agreements but rather provide only essential information necessary for an individual to make a decision More detailed disclosures (such as FAQs in plain language) should be made easily accessible to those who wish to learn more with no hidden surprises
bull An opt-in approach to authorization should be instituted to ac-company initial DCTT rollout The feasibility and value of opt-out approaches should continue to be evaluated informed by what is technologically possible what local assessments of benefits and harms of the technology reveal over time and our evolving un-derstanding of the degree to which an opt-out approach is likely to increase or decrease utilization among different populations Opt-out approaches should not be precluded
Promoting Equity and Fairness in Application of DCTT
bull States localities and institutions that recommend widespread use of DCTT should provide technology (eg mobile phones Blue-tooth devices) and free data packages to those who desire but lack access to these devices
bull If there are lower rates of adoption of DCTT systems in some iden-tifiable communities public health authorities should find ways to compensate For example directing more non-DCTT resources and efforts toward those communities to meet specific needs that are elsewhere being supported by technology
bull If maps are generated based on DCTT to provide the public with the locations that COV+ individuals have visited steps must be taken to minimize the stigma and potential financial losses that could result from a location being identified as a hotspot
Instituting Transparent Governance and Oversight
bull Digital surveillance oversight committees should be established expeditiously with diverse and qualified membership to provide ethical and regulatory review prior to and concurrent with wide-spread use of a DCTT system
bull Understandable and publicly accessible rules must guide the col-
12 Digital Contact Tracing for Pandemic Response
lection access control use storage and combination of data by government authorities public and private institutions and other parties such as public health researchers
bull Only those data that are necessary and relevant for the public health response to COVID-19 should be collected and used
bull Identifiable data should be kept only for the period of time needed for the public health response to COVID-19
bull Identifiable data collected as part of this response should not be shared with anyone other than the relevant public health authorities without additional specific informed consent of individual users
bull Before a government or institution adopts a digital contact tracing program they should state the conditions under which the digital contact tracing program will be terminated
bull Future use of DCTT to advance public health or other efforts (eg use in seasonal flu surveillance) would require independent justification DCTT designed for public health use should not be used by law or immigration enforcement
bull The principles offered in this guidance document apply both during and following the COVID-19 pandemic
Legislative Recommendations
bull The United States Congress should enact new legislation specif-ically tailored to facilitate the use of DCTT as part of the public health response to COVID-19 while also protecting user privacy and ensuring data security
bull Congress should require DCTT developers to disclose to users in clear language the nature of the information that would be col-lected how it would be collected how it would be stored and for what purposes it may be used
bull While the rollout of DCTT should initially employ an opt-in au-thorization approach the feasibility acceptability and value of opt-out approaches should continue to be evaluated As such opt-out approaches to consent should not be precluded by legislation
Summary 13
bull Congress should prohibit the commercial use of data collected for COVID-19 response by DCTT
bull Congress should prohibit discrimination on the basis of data col-lected by DCTT
bull If Congress is unable to enact suitable legislation state legislatures should work toward enacting similar laws for their jurisdictions A ldquomodelrdquo state law should be rapidly developed to facilitate na-tionwide adoption of an appropriate law and uniformity of legal requirements
Summary of Analysis
Supporting Effective and Informed Adoption
The COVID-19 pandemic and the physical distancing efforts imple-mented to slow the rate of transmission have caused severe harm to indi-viduals communities and our society To protect the public good going forward we need a robust public health response that reduces the spread of SARS-CoV-2 and does so in a way that allows economic recovery to occur and to be sustained We also need to design and manage this public health response so as to minimize harms to individuals and society to distribute benefits and burdens equitably across the population and to avoid misuses of the technologies and the data they collect
To reduce the spread of SARS-CoV-2 chains of transmission need to be broken To do this people who have been exposed to SARS-CoV-2 or potentially exposed need to be identified as comprehensively and as quickly as possible so they can quarantine themselves and avoid infecting others This is the job of manual contact tracing by public health authori-ties in which people infected or presumptively infected with SARS-CoV-2 are interviewed and asked about their movements and interactions in-cluding where they work and shop how they travel with whom theyrsquove had contact and the nature of that contact (eg where the contact took place) Their contacts are then interviewed and potentially asked to quar-antine seek testing and take other protective measures if the contact is sufficiently high risk
14 Digital Contact Tracing for Pandemic Response
The hope is that DCTT can augment traditional contact tracing ef-forts either by working alongside and independently of manual contact tracing or by being integrated into manual contact tracing efforts in a way that makes these efforts faster more thorough and more efficient
Data suggest that a substantial proportion of transmissionsmdashper-haps as high as 50mdashoccur between individuals who are not symptom-atic and that transmission may occur as early as 3 days before onset of symptoms (WHO 2020) Because asymptomatic spread of SARS-CoV-2 appears to be a significant source of infection we need to identify po-tentially infected people before they show symptoms thus speed is of the essence This is one benefit of using DCTT potential contacts can be identified instantaneously notified quickly and asked to quarantine as soon as possible
Another benefit is identifying contacts who manual contact tracing methods may miss either because COV+ people do not remember all the places theyrsquove been or cannot identify all the people theyrsquove had contact with This is especially relevant given the long period of infectivity of SARS-CoV-2 which begins before people are symptomatic and aware they are infected (Ferretti et al 2020) If DCTT were designed to have optional location-monitoring capabilities this critical challenge could be mitigated even further For example location data might reveal that a COV+ person was at a restaurant at an exact time and date which could be followed up by contact tracers who could alert the public or use other measures to reach those who were also present in the restaurant at the same time In other disease contexts geolocation data have demonstrated some potential to support epidemiology and disease surveillance (see Fur-lanello et al 2002 Dredze et al 2013 Eckhoff and Tatem 2015 Fraccaro et al 2019) with technical cautions regarding accuracy and the like (Beu-kenhorst et al 2017)
One role for DCTT is to work alongside manual contact tracing but independently of it Individuals would download proximity tracing or exposure notification apps use them receive alerts if theyrsquove had a poten-tial contact with another user who is COV+ or presumptively COV+ and voluntarily self-quarantine without having contact with public health au-thorities or giving them data that feeds into public health contact tracing efforts It is possible that this would help to break chains of transmission and reduce the spread of SARS-CoV-2 though at this point these benefits
Summary 15
are speculative It is also possible that such exposure notifications will result in high rates of false positives
Another possible role is for DCTT to be integrated into manual con-tact tracing efforts When potential contacts are identified by DCTT they are connected to public health authorities who can then follow up with them There are different forms this could take and different kinds and amounts of data about contacts public health authorities could receive from DCTT On one end of the spectrum of reporting public health authorities would not receive individualsrsquo names or contact information only anonymous data The fullest version of reporting would securely send to public health authorities the names contact information such as address and phone number and other data about contacts that DCTT collected including data about their location and movement history
It is uncertain whether providing public health authorities with vol-umes of information on cases and contacts from DCTT will be useful in practice As mentioned above providing public health authorities with location data on cases and contacts collected by DCTT may help con-tact tracers to find and notify additional contacts However at present providing public health authorities with large amounts of data will be useful only if there is sufficient capacity to follow up on these data In addition there is a risk of low-quality data from DCTT flooding the system leading to investigation of false case contacts identified by DCTT and distracting from other important efforts Whether and to what extent data from DCTT will benefit contact tracing efforts is unknown pointing again to the importance of continuously collecting high-quality evidence about DCTT
Designing Flexible Technology to Maximize
Public Health Utility While Respecting Other Values
Use of DCTT is essentially an experiment as we have insufficient infor-mation about the performance of different DCTT and their efficacy In the face of this uncertainty how should DCTT be designed and how should its use be managed
Many efforts to advance DCTT in the United States and elsewhere have emphasized the importance of ldquoprivacy by designrdquo that is building privacy and security protections into the design of technology rather than counting on responsible use alone (Cavoukian 2010) As noted above
16 Digital Contact Tracing for Pandemic Response
some major technology companies have signaled this position through de-velopment of PPPT systems that embed features such as decentralization de-identified information user anonymity bans on collection of location data and minimal reliance on or integration of public health authorities or other government actors Many of these features have also been em-braced early by advocacy organizations (Crocker Opsahl and Cyphers 2020 Electronic Privacy Information Center 2020 Kahn Gilmor 2020) and in an open letter (ldquoJoint Statement on Contact Tracingrdquo 2020) from nearly 300 researchers These same actors have emphasized that use of DCTT should be fully voluntary
Although privacy is a key value individuals and communities may also value efficiency equity liberty autonomy economic well-being com-panionship patriotism or solidarity among other values People may accept more significant encroachments on privacy now if this ultimately results in realizing other values (such as companionship) that are of equal or greater importance to those individuals Rather than centering pri-vacy alone in design a different orientation is needed at this moment that of ldquovalues in designrdquo which incorporates a broader range of values into technology (Flanagan Howe and Nissenbaum 2008 Knobel and Bowker 2011) For example some users might wish to express autonomy solidarity or patriotism through DCTT by sharing their location history with public health professionals in order to advance the public health re-sponse increase system efficiencies (eg by contributing information that can lead to better data processing) and reduce the burden on essential workers At the same time there is value in further advancing autonomy by designing technology to allow individuals some control over what data about them are collected and shared
DCTT should be designed to have a base set of features that protect privacy and strive for interoperability but also should include other op-tional capabilities This could be achieved by designing DCTT to have a default that can be modified for example an initial setting could be that usersrsquo location data are not shared with public health authorities but us-ers may opt-in to this feature Such an opt-in approach is likely consistent with existing federal privacy laws
Designing DCTT this way gives users the flexibility to decide how to use the technology and how to engage with public health authori-ties consistent with their values and trade-offs they are willing to make This flexibility could also allow for more real-world evaluation of how
Summary 17
different users experience different features of DCTT in different loca-tions Technology design should not be static but it should be capable of evolving depending upon local conditions new evidence and changing preferences and priorities
DCTT developers must comply with a number of federal privacy laws These privacy laws generally permit the collection storage and use of personal information so long as the user provides meaningful consent Privacy law in the United States is generally sector-specific and limited in scope resulting in a patchwork of protections that differ significantly depending on the entity that collects the data and the type of data col-lected Given the complexity of existing federal privacy law and the need to further strengthen public trust in DCTT it would be beneficial for Congress to enact new privacy legislation that is specifically tailored to the use of DCTT in response to COVID-19 Such COVID-specific legisla-tion should be sensitive to the full range of values and recommendations described above
In short designing ldquomiddle-groundrdquo DCTT for flexible use may pro-vide the most adaptable and thus most robust public health responsemdashrespecting privacy and individual autonomy by allowing users to use DCTT in ways that express their own values
Public Acceptance of DCTT
While some groups have maintained that only PPPT-like minimal systems will be widely adopted because only they will earn and maintain public trust (Simpson and Conner 2020) there is insufficient evidence that pub-lic trust would be threatened by a DCTT system that has the capacity to securely collect location data integrate public health authorities and en-able voluntary sharing of certain user data (eg location data) with those authorities More research including through deliberative engagement sessions is needed to better understand how differences in the features and functionality of DCTT (such as optional sharing of geolocation data) influence trust and peoplersquos willingness to use DCTT Technology com-panies should not alone control the terms conditions and capabilities of DCTT nor should they presume to know what is acceptable to members of the public
Significant concerns have also been expressed by privacy advocates (Guariglia 2020) and in the popular press (Giglio 2020) about ldquosurveil-lance creeprdquomdashthat is a belief that state or corporate actors will use new
18 Digital Contact Tracing for Pandemic Response
surveillance technologies capacities and permissions well beyond the purposes for which they were initially justified to the public and beyond the time when they are useful for the COVID-19 pandemic Surveillance creep is a serious concern and should be carefully guarded against how-ever the possibility of surveillance creep is not a sufficient reason to limit development of DCTT to minimal systems Instead protections should be put in place to ensure that only those data that are necessary and relevant for the public health purposes at hand are collected and used and data should be kept only for the period of time needed for those pub-lic health purposes For this reason we would support COVID-specific legislation that would impose strict limits on the use of DCTT data for nonndashpublic health purposes
Finally the use of DCTT during the current pandemic should not set a precedent for future public health use (eg use in seasonal flu surveil-lance efforts) Future use would require independent justification Further use of DCTT in other contexts (eg by law enforcement or immigration enforcement) is presumptively unethical
Encouraging Adoption of DCTT
Researchers have estimated perhaps conservatively that DCTT use by 80 of smartphone owners or 56 of the population overall will be needed to suppress the epidemic (Hinch et al 2020) These estimates also highlight that some decrease in transmission would be realizable even with lower rates of technology adoption
In the United States many advocates and researchers have argued that use of DCTT must be fully voluntary However experience from other countries suggests that when use of a digital contact tracing app is voluntary only a minority of the population will download it Instead of making use fully voluntary and initiated by users there are ways that DCTT could be put into use without usersrsquo voluntary choice For exam-ple use of an app could be mandated as a precondition for returning to work or school or even further to control entry into a facility or trans-portation (such as airplanes) through scanning of a QR code to demon-strate personal exposure levels (Gan and Culver 2020)
While these approaches are hard to imagine in the United States some have argued that mandatory use of DCTT could be ethical If man-dates increase adoption of DCTT and improve the public health response
Summary 19
this would reduce the likelihood of lockdowns which are harmful and a severe limitation of individual liberty applied on a mass scale On the other hand mandated use of DCTT systems may not be effective People may not adhere to the mandate by simply leaving their phone at home Perhaps more important should the technology not deliver the hoped-for benefits having mandated the use of an unproven technology could result in a loss of public trust in the technology in the entity instituting the mandate and in the larger public health response potentially lead-ing to noncompliance with public health recommendations more broadly (Bernstein et al 2019)
Any decision maker considering mandatory use including govern-ment officials institutional leaders and employers must convincingly address a number of considerations Particularly important is the need to identify reliable evidence that the DCTT would be effective and to ensure that the burdens and benefits of use are equitable and justifiable At this time mandated use of DCTT by states or institutions is not jus-tifiable given uncertainty about potential harms and benefits Additional technology user and real-world testing is needed before mandatory use should be considered
As with any public health effort the amount of evidence that must be offered to illustrate that the intervention or program can achieve its aims and the degree to which people should be able to exercise choice in their participation should be in proportion to the anticipated bur-dens of the intervention or program For example the permissibility of mandating use of DCTT by the public depends on factors such as the sensitivity of the data that are collected the extent to which public health is integrated within the DCTT system and what actions are taken in response to confirmed virus exposure or being identified as COV+ (eg forced quarantine) The more burdens that are placed on individualsmdashfor example whether people are ordered into quarantine if they have been exposed to the virus or if there are limited social supports for those in quarantinemdashthe greater the demand should be on the performance of the DCTT system
Perhaps the most effective way to generate widespread US adoption of DCTT will be to offer incentives for its use in other contexts gener-ally speaking small incentives have been shown to lead to an increase in desired outcome (Singer and Ye 2013 Lee et al 2014) Given the impor-
20 Digital Contact Tracing for Pandemic Response
tance of widespread use modest incentives ought to be considered if and when there is sufficient evidence of the utility of DCTT so long as those incentives are not mandates in disguise Another ldquofirst linerdquo approach to increasing use of DCTT is for trusted community leaders public figures health care professionals and other respected individuals to communicate with the public and their communities about DCTT and to encourage its use through public engagement campaigns if and when the technology demonstrates sufficient potential
Establishing Meaningful Processes for
User Disclosure and Authorization (Consent)
Any effort to roll out DCTT should ensure that users have a meaning-ful opportunity to review and understand information about the specific technology and its uses Moreover given the importance of public trust and the current crisis of public trust in governments and technology com-panies handling private digital information there is a strong ethics argu-ment for requiring consent from individual users We recommend a care-fully crafted version of what is sometimes called simple consent which consists of basic disclosure and voluntary agreement or authorization This disclosure should include information about the purposes of the technology the userrsquos options for collecting and sharing data purposes for which data can be used and any known risks among other informa-tion This information should be presented in an accessible format on any DCTT app and more detailed disclosures should be readily accessible for those who wish to review them
Through an opt-in mechanism such as clicking a button to signal agreement users should be able to indicate their intention to use a DCTT The opt-in approach is consistent with mechanisms for agreement to use other downloaded applications An opt-in approach should be part of the initial introduction of DCTT given the novelty of the technology and its uses and the need to build trust and confidence in the system Successes of opt-out approaches in other areas suggest that the feasibility and value of an opt-out approach to DCTT should be carefully evaluated particularly in conjunction with assessment of whether public health goals are being met (Rithalia et al 2009) Such assessments should be informed by what is technologically possible by local data regarding benefits and harms of the technology and by evolving understanding of the degree to which
Summary 21
an opt-out approach is likely to increase or decrease utilization among different populations
Promoting Equity and Fairness in Application of DCTT
Digital contact tracing technology should be designed and used in ways that as far as possible promote an equitable distribution of benefits and burdens DCTT should be deployed in a manner that does not propagate preexisting patterns of unfair disadvantage or distribute harms and risks unfairly throughout the population It is well known that some commu-nities have lower rates of technology and data access and therefore may benefit less from use of DCTT unless steps are taken to address these digital disparities Additionally should use of DCTT be made a require-ment for entry into a workplace into a school or onto transportation then those who currently do not possess the required technology must not be unfairly burdened through lack of access In order to mitigate this states localities and institutions that recommend widespread use of DCTT should provide technology (eg mobile phones Bluetooth de-vices) and free data packages to members of the community who desire but lack access to these devices
Some populations may also experience greater harm and greater fear of harm from having their data collected For example some groups such as African Americans Hispanic Americans Muslim Americans and undocumented immigrants have more reasonable fear of their data being handed over to law or immigration enforcement and some groups have lower levels of trust in public health due to past injustices (Auxier et al 2019 CSM 2017 Pew Research Center 2017 Rodrigues et al 2018) This further substantiates the need to limit use of any data gathered by DCTT to its public health purpose
Instituting Transparent Governance and Oversight
DCTT must be developed with an eye toward both present and future implications We are rapidly gaining knowledge about SARS-CoV-2 and COVID-19 but still have essential gaps in our understanding In the United States public health responses including DCTT will generally be developed and coordinated by individual states regional consortia (Res-ton Sgueglia and Mossburg 2020) and associations Good governance in this context requires transparency and the creation of oversight bodies
22 Digital Contact Tracing for Pandemic Response
with the appropriate expertise and representation to allow nimble and effective responses while serving as trusted representatives
In order to address the range of ethics and governance concerns that relate to the design and use of DCTT we recommend that digital surveil-lance oversight committees be established perhaps at a state level and with a platform for national coordination These committees can provide ethics and regulatory review prior to and concurrent with widespread use of DCTT The committees should be composed of a diverse group of experts capable of evaluating the quality of a DCTT system locally including members of communities that experience higher rates of digital disparity
When assessing the design and use of digital contact tracing systems these committees (and the public more widely) should consider not only the risks and benefits accrued during the COVID-19 pandemic but also implications for the future How can we navigate safe use of these tech-nologies in a way that preserves public trust in them and enables the possibility of future beneficial use
As a start it should be emphasized that the principles offered in this and other guidance documents do not apply only during the pandemic Future efforts to advance DCTT capabilities during quieter times should make every effort to follow them
23
Introduction
Public health professionals around the world are working tirelessly to respond to the COVID-19 pandemic using tried-and-tested public health methods for infectious disease surveillance and control These traditional methods are essential to the global COVID-19 response To complement these actions and potentially augment the speed and efficacy of the public health workforce digital technologies are being harnessed Given the scale of the pandemic significant efforts are being undertaken to develop and leverage public-facing and health-system-supportive tech-nology solutions including smartphone apps and other digital tools that may aid public health surveillance and contact tracing
Digital contact tracing technology and closely related digital health products (hereafter DCTT) have been used in several countries as part of broader disease surveillance and containment strategies Globally many digital COVID-19 contact tracing strategies have already emerged in response to the pandemic This is not surprising given the ubiquity of mobile phones and other digital devices around the world (ldquoMobile Cellular Subscriptions (per 100 People)rdquo 2018) experiences developed during prior outbreaks and pandemics and the pre-COVID-19 momen-tum behind using digital technologies to support individual and health system capabilities (WHO 2017 Mathews et al 2019 Aiello Renson and Zivich 2020 Mahmood et al 2020) In the United States DCTT has been proposed as an integral part of some plans to ldquoreopenrdquo the country (Allen et al 2020 Hart et al 2020 Simpson and Conner 2020) It is al-most certain that these and related technologies will become part of not only the COVID-19 response but also the larger toolbox for future public health communicable disease prevention and control
24 Digital Contact Tracing for Pandemic Response
While novel public health surveillance technologies such as DCTT have theoretical promise their effectiveness is unclear These technologies also raise important ethical legal and governance challenges that require comprehensive analysis in order to support decision-making regarding their appropriate use A number of frameworks recommendations and analyses have emerged recently in an effort to chart potentially ldquosaferdquo pathways for use of public health disease surveillance technology Many in the United States such as the Electronic Frontier Foundation Elec-tronic Privacy Information Center American Civil Liberties Union and the Center for American Progress are proposing that digital public health surveillance technologies must embrace strict data privacy protections decentralized data storage a high degree of anonymity and voluntary adoption (Crocker Opsahl and Cyphers 2020 Electronic Privacy Infor-mation Center 2020 Kahn Gilmor 2020 Simpson and Conner 2020) Others have argued that technologies that seek to enhance public health response during a pandemic should more closely align with the needs of public health professionals and the evidence-based procedures they follow stating that interests in serving the publicrsquos health ought to weigh more heavily in the necessary balancing of stakeholder interests (de Jong et al 2019 Watson et al 2020) This view is in part based on a recogni-tion that during countless other outbreaks the public has benefited from traditional disease surveillance and contact tracing which are heavily re-liant on centralized data storage and when necessary the collection of identifiable information These traditional approaches are governed by ethics principles (PHLS 2002) ethics guidelines (WHO 2017) and laws (ASTHO 2012) and digital technologies represent a new tool to support them
While debates and recommendations about appropriate design and use of DCTT have focused intensely on minimizing important data-related risks a wider lens is needed to fully appreciate the many additional criti-cal questions that need attention This report begins to grapple with these questions which are critical to address in order to guide responsible use of DCTT Given the complexity of the terrain as a first step toward estab-lishing a foundation for responsible decision-making regarding potential use of DCTT we offer a set of guiding principles (see box) These prin-ciples are meant to apply to DCTT as well as other digital technologies used in novel ways during pandemic response
Guiding Principles for the Use of Digital Public Health Technologies for Pandemic Response
Transparency and public engagement are essential to an inclusive digital public health response
bull Government public health and digital technology leaders must engage
effectively with the public and other stakeholders to communicate the util-
ity importance oversight and limitations of relevant digital technologies
including their implications for individualsrsquo privacy and civil liberties
bull Transparency at all levels is essential for maintaining public trust and
confidence
bull To the extent possible digital public health responses should reflect the
range of values that are important to individuals including advancing the
health and well-being of the community as a whole
bull Decision makers should recognize the sacrifices that some people may be
willing to make during a pandemic in order to advance public health goals
Acceptance by some of particular monitoring capabilities should not be read
as a willingness to extend these methods to other problems or uses
Digital public health responses must represent the least infringement of civil liberties necessary to accomplish the public health goals
bull If preferred digital public health strategies infringe on privacy and other civil
liberties the infringements must be sufficiently justified by the circum-
stances of the pandemic offset by ample anticipated public benefit and
considered relative to infringements associated with other possible strate-
gies such as mass physical distancing
bull Only those data that are necessary and relevant for the stated public health
purposes should be collected Identifiable data should be stored in a secure
manner and only for the period of time that the public health purposes
require
bull Adopted technologies should not be used in ways that subject communities
to discrimination or surveillance for nonndashpublic health reasons
bull Respect for individual autonomy requires that users are sufficiently informed
of the public health goals of the technology and the extent to which those
goals are being met
Use of digital public health technologies and data must be guided by best available evidence
bull Decisions to deploy digital public health technologies should be based on a
careful assessment of the uses and limitations of any proposed technology
taking into account the best available evidence
bull Those who deploy digital public health technologies should continuously and
systematically monitor their performance as well as any evidence that is being
generated in other contexts about the selected technological solution and about
other competing technologies
bull Unintended consequencesmdashincluding those that might impact public health
goals core values and interests of the public and unfair advantage or disadvan-
tagemdashshould be carefully monitored and addressed as necessary
Responsible use of digital public health technology requires meaningful governance and accountability
bull Systems of governance must be trustworthy and well informed They must be
reviewed and adjusted as circumstances and evidence change or as unintended
effects are identified
bull Trusted representatives who are capable of developing and implementing uni-
form and fair standards for adopting and utilizing underlying digital technology
must be identified
bull Understandable transparent and publicly accessible rules must guide the
collection access control use storage and combination of data by government
authorities public and private institutions and other parties such as public
health researchers
bull Oversight accountability and consequences for abuse or misuse of these data
must be explicit and enforceable
The deployment of digital public health technology must be rooted in a commitment to equity
bull Digital public health technologies should be deployed in a manner that does
not propagate preexisting patterns of unfair disadvantage or further distribute
harms and risks unfairly throughout the population
bull To the extent possible digital public health technologies should be designed to
rectify existing inequities
bull Oversight mechanisms must be in place to ensure that the improved public
health outcomes are equitable and to detect and correct any unforeseen resul-
tant injustices attributable to the technology or that can be addressed using the
technology
bull The incentives and disincentives for adopting new technology must be equita-
ble not exploitative and aligned with effective use of the technology
bull Disparity-driven technology gaps should be explicitly recognized To the extent
possible provisions should be made to address the digital divide
Introduction 27
In reflecting on these principles it becomes clear that if we wish not only to realize but to maximize the public good that might come from use of DCTT we must carefully define and responsibly align public health needs and capabilities with technological needs and capabilities We must understand that although technology may serve as a workforce multiplier it alone will not solve the public health challenges we face We must identify and address assumptions and misinformation about technologies and data use We must provide the means and opportu-nity for informed decision-making by the public and those who serve as our representatives Government officials public health leaders leaders of other institutions employers digital technology developers and the public all must be adequately informed and engaged in order to make the best decisions possible under the circumstances
ONE
29
Types of Information Collected through Contact Tracing
Data Collected from Infected Persons
Symptoms and Course of Illness
Information about COVID-19 patientsrsquo signs symptoms and course of illness is important to public health because it provides a basis for refining clinical case definitions and informing health care providers and the gen-eral public (CDC 2020c) This includes the specific signs and symptoms manifested by persons who are COV+ as well as the relative frequency and durations of different signs and symptoms This would also take into consideration those persons with no symptoms but who test positivemdashthose who are presymptomatic (develop symptoms later) those who are postsymptomatic (clinically recovered but still infectious) and those who never manifest illness at all
Typically contact tracing begins with a case in which a person has confirmation of infection by means of a diagnostic test However in some cases test results are not reported until several days later and individuals may be identified as ldquopresumptive positiverdquo cases until testing can be completed In these cases contact tracing efforts will need to be updated when test results are returned For example if a test comes back negative public health professionals will want to notify contacts that they no lon-ger need to quarantine
Public Health Perspective
30 Digital Contact Tracing for Pandemic Response
Movement and Contacts
In order to manage cases appropriately (identify and track the infected isolate the sick quarantine the exposed) public health officials need de-tails on each case (Resolve to Save Lives nd) First they need to know who and where the individual is That means personally identifiable infor-mation and contact information (address phone numbers email) It also means information about the nature intensity and duration of contact with individuals to whom they may have transmitted the disease This may include information about where the individual works and the kind of work they do (eg health care worker) how they travel (eg bus sub-way car) and where they shop or any other public venues they may have visited during a period of possible infectiousness (PIH 2020a) It may be helpful in certain circumstances for public health officials to ensure that suspected cases contacts or other high-risk individuals are following iso-lation and quarantine recommendations or orders
Contact tracing involves identifying all individuals who have had sig-nificant exposure to confirmed or probable cases during the time prior to and after the onset of symptoms both of which are times when the case is thought to be infectious (Africa CDC 2020) Contacts could be those who are caring for COVID-19 patients especially if they lacked proper PPE and those who had close interaction with the COV+ person over a sustained period of time particularly in enclosed spaces (PIH 2020a) For COVID-19 contacts are identified by asking a person with a confirmed or probable case about people they may have been within 6 feet of for 15 minutes or more starting from 48 hours before the onset of symptoms and lasting until the person is isolated (CDC 2020b)
Data Collected from Contacts of Infected Persons
Contact Details
In addition to the data collected from individuals with COVID-19 con-tact tracers will collect data from potentially exposed individuals (con-tacts) Information about the nature intensity and duration of contact with an infected person may be collected for a contact if information about the case is known to the contact These details can help a contact tracer more accurately determine whether the contact is at high or low risk for SARS-CoV-2 transmission and help determine whether a con-tact should quarantine for 14 days (the upper bound of the SARS-CoV-2
Public Health Perspective 31
incubation period) In addition public health professionals may gather contactsrsquo demographic information and other personal data to contribute to population-level disease surveillance and situational awareness about an epidemic (CDC 2005) However the information needed at baseline is only a personrsquos name and contact information
Symptoms (If They Develop) and Course of Illness as well as Information about Close Contacts
If a contact develops COVID-19 symptoms while in quarantine andor tests positive for the virus public health will then collect the data required for a COVID-19 case This includes collecting information on the con-tacts that a person may have had (if any) in the days immediately before and during the course of their infection
How Contact Tracing Information Informs Public Health Action
To reduce disease burden and help make ldquoreopeningrdquo safer during the COVID-19 pandemic the United States and other countries will need to identify gather information about and safely isolate cases and quar-antine their contacts to reduce community transmission (Watson et al 2020) Gathering information about possible cases and their contacts en-ables public health to break chains of transmission
Contact tracing involves stages (CDC 2020a) including
1 identifying an infected person as a COVID-19 case
2 identifying the close contacts of that case (Africa CDC 2020)
3 getting in touch with contacts
4 asking contacts to quarantine at home for 14 days
5 assessing contacts for possible symptoms and
6 following up with COV+ persons and their contacts to identify new or worsening symptoms and connect them with medical care if needed
Contact tracers also play an important role in providing resources for COV+ persons who are in home isolation and their contacts who are
32 Digital Contact Tracing for Pandemic Response
in home quarantine Knowing who and where cases and contacts are can enable provision of supplies such as digital thermometers or masks Effective contact tracing that enables isolated cases and quarantined con-tacts to remain at home also requires providing a range of social sup-port services or ldquocare packagesrdquo from delivering food and medicines to trash pickup Furthermore vulnerable individuals who are homeless or otherwise unable to sufficiently isolate or quarantine in their current living conditions may need to have alternative housing arranged to safely remain separated from others (CDC 2020b)
Finally contact tracers explain what quarantined contacts should do if they begin to develop symptoms consistent with COVID-19 (Africa CDC 2020) Depending on the context contact tracers may engage in active monitoring by regularly communicating with contacts about their health status through phone text message or possibly mobile applica-tions In rare cases public health can make quarantine mandatory and may monitor a quarantined individual to ensure that they do not break quarantine Contact tracers may also facilitate access to health care by providing telemedicine resources or other information and support for accessing medical care
Characteristics That Make Data Useful to Public Health for Reducing Disease Transmission
Data Access
If digital contact tracing technology and closely related digital health products (together DCTT) are intended to support the public health ac-tions described above and directly amplify public health capacity to con-duct case identification and contact tracing then data collected through DCTT must be accessible to public health authorities Identifying infor-mation and location data for cases and contacts of cases are necessary for public health use so that contact tracers can do their work to uncover ongoing transmission and enable isolation and quarantine These data should also be durable meaning that public health can return to the data in order to interact with and support cases and contacts These data can also be useful at a population level if de-identified and aggregated by illuminating trends in community transmission and providing support for decisions about resource allocation
Public Health Perspective 33
Data Format
Data should be provided to public health authorities in a usable format that is compatible with public health systems and that has the granularity and specificity of personal information that is needed for use in contact tracing Without personal identifiers the data cannot be used by public health workers to undertake contact tracing Data should also contain information about the nature of a contact including the proximity of the contact and number of minutes that the person was in contact with an infected individual Location data can also help public health author-ities to conduct contact tracing particularly when contact occurred in a crowded area and involved people who donrsquot know one another Loca-tion data from a case can help public health professionals identify con-tacts even when those contacts themselves are not using a contact tracing app because the data shows contact tracers where to look for additional contacts
Data Accuracy
Data that identifies individuals as having sustained contact with a case must be as accurate as possible If criteria for being considered a contact are too restrictive it may result in missed contacts and sustained chains of disease transmission If criteria are too broad it may result in unneces-sary restriction of movement which could have significant personal and economic consequences
Timeliness of Data
Data from cases and contacts must be timely in order to enable case-based management that will help reduce community transmission For contact tracing to be effective infected individuals need to be isolated and their contacts identified and quarantined as quickly as possible Testing for SARS-CoV-2 can take time sometimes many days for a test result Especially because SARS-CoV-2 is transmissible during the pre-symptomatic period data on symptomatic individuals should be made available to public health officials even before a positive test is returned in order to enable identification and quarantine of contacts right away If this information is delayed until a test result is received it may be too late to identify and quarantine contacts because contacts (if infected) will already be contagious and may have spread the virus to others
34 Digital Contact Tracing for Pandemic Response
VolumeAvailability of Data
The more that individuals opt to share their information to support con-tact tracing the more effective contact tracing will be in breaking chains of viral transmission and controlling epidemics of COVID-19 The exact proportion of cases and contacts that need to be identified in order to avoid large surges of cases which overwhelm health care systems is un-certain but the goal is to identify all infected cases and all close contacts of each case (PIH 2020b)
Recommendations
bull Technologies or apps with the goal of enhancing public health ca-pacity to identify cases and trace contacts in order to control the spread of SARS-CoV-2 should be designed to match functionality with that goal
bull Technologies or apps may produce some false negatives or false positives but they should be accurate enough that public health authorities feel confident that they support and donrsquot detract from contact tracing efforts
bull DCTT approaches for public health should be designed to facilitate the following
deg identifying contacts including those who may not be easily found otherwise
deg finding and notifying contacts rapidly before they develop symptoms if infected
deg analyzing the nature of contact to determine whether contact is high medium or low risk and to support decisions about whether a contact should quarantine and
deg following up with cases and contacts so that public health can provide resources to support isolation and quarantine
bull Data collected through DCTT should be made available to public health professionals and to researchers in de-identified form to sup-port population-level epidemiologic analysis
T WO
35
Characteristics of SARS-CoV-2 Relevant to Candidate Digital Solutions
The SARS-CoV-2 virus has some unique transmission characteristics and clinical manifestations that can help guide use of digital contact trac-ing solutions Individuals infected with this virus may or may not show symptoms or may show a range of different and sometimes nonspecific symptoms Estimates regarding the percentage of individuals who are in-fected but never develop symptoms is highly uncertain ranging from 5 to 50 (Heneghan Brassey and Jefferson 2020) Data suggest that a substantial proportion of transmissionsmdashperhaps as high as 50mdashoccur between individuals who are not symptomatic and that transmissibility may extend out as long as 3 days before the onset of symptoms (WHO 2020)
The complexity of asymptomatic and presymptomatic transmission makes it more difficult to identify all cases of COVID-19 It also means that manual contact tracing is less effective because people are unlikely to remember all of their contacts during the long period of infectivity (Ferretti et al 2020) however it does not negate the need for contact tracing Identifying symptomatic cases will still help greatly with slowing the spread because their contacts can be asked to quarantine to prevent them from spreading the virus if they are indeed infected This means that whether they become symptomatic or not contacts will be quarantined and the chain of transmission will be broken If contact tracing can be implemented on a large enough scale perhaps with support from DCTT
Digital Technology and Contact Tracing
36 Digital Contact Tracing for Pandemic Response
eventually the virus could be managed at much lower levels of community transmission and large epidemics of unrecognized spread will not occur
The transmissibility of the virus when a person has no symptoms further suggests that effective solutions may require multimodal inter-ventions combining contact tracing with frequent rapid and ubiquitous testing and continued social distancing to varying extents (Cheng et al 2020)
Because of presymptomatic spread contact tracing efforts and dig-ital solutions to augment those efforts should support identification of contacts a person had 2 days before their symptoms and at least 3 days after the resolution of those symptoms (if the person continued to have contacts through that time period) (CDC 2020d) Additionally public health messages delivered by these technologies should urge contacts to quarantine for the full 14-day incubation period
Previously Existing Contact Tracing Technologies
Prior to this pandemic health agencies in high- medium- and low-income countries had begun to develop and use digital tools to augment the man-agement of infectious diseases including sexually transmitted infections (HIV chlamydia gonorrhea) and high-consequence epidemics (Ebola) (Danquah et al 2019)) However these have been primarily used to facil-itate case interviews partner notification (in the case of STIs) and record keeping as opposed to fully digitizing or automating the contact tracing process
It has been recently suggested that digital contact tracing could con-tribute to the management of the ongoing COVID-19 pandemic and the experiences of containing SARS-CoV-2 in countries such as China Sin-gapore and South Korea provide noteworthy examples However un-dertaking this case-based intervention on the scale required to achieve pandemic control is a novelty in the history of public health Although technological development is proceeding rapidly several foundational is-sues have yet to be resolved including functionality connectivity to pub-lic health authorities and informatics systems usability by disease inter-vention specialists (DIS also referred to as contact tracers) and sufficient protection of personally identifiable information among others
Digital Technology and Contact Tracing 37
Introduction of Novel Digital Contact Tracing Technologies
Digital contact tracing technologies and platforms have recently been in-troduced and the CDC has published preliminary criteria for evaluating these tools (CDC 2020e) It can be helpful to consider three broad ap-proaches along the spectrum of potential methods of digital contact trac-ing a maximal approach (typified by the South Korean governmentrsquos cen-tralized data collection (M S Kim 2020)) a minimal approach (typified by the AppleGoogle decentralized privacy-protecting proximity tracking (Apple and Google nd)) and a diverse middle ground that aims to aug-ment manual contact tracing with the collection of digital data Perhaps the most promising approach in this middle ground involves allowing us-ers to turn over both proximity data and GPS location data (ie cell-site location data) to public health authorities on a voluntary basis
Along with this ldquominimal to maximalrdquo spectrum in the design of dig-ital contact tracing technologies and systems there is another spectrum that concerns voluntary versus mandatory use of these technologies are individuals entirely free to use these technologies or not or should poli-cies incentivize or even mandate their use At one extreme South Korea (Republic of Korea) implemented a system (called Safe Korea) supported by the Ministry of the Interior and Safety that collects a variety of per-sonal data in a centralized database in order to enforce quarantine orders and track possible contacts (M S Kim 2020) Israel also implemented a centralized involuntary data collection system for tracking COVID-19 cases and alerting those who may have been exposed (Hendrix and Eg-lash 2020) In Poland health authorities have set up mandatory ldquocheck-insrdquo involving a GPS-waypoint capture and ldquoselfierdquo photographs sent to the monitoring agency to ensure that individuals are not breaking quar-antine (Hamilton 2020)
These centralized systems can be designed to incorporate data from a variety of sources The data collected include location data from mobile phones QR codes can also be scanned to track the use of public transit where GPS data may be inadequate (due to low resolution) to accurately distinguish the occupants of one vehicle from another The data collected from mobile phones can then be integrated with data from other sources such as facial-recognition cameras credit card transactions and social media
38 Digital Contact Tracing for Pandemic Response
At the other extreme of technology invasiveness for contact trac-ing isolation and quarantine many corporations and working groups (including the AppleGoogle collaboration) have developed privacy-pre-serving proximity tracking (PPPT) using Bluetooth Low Energy (BLE) ldquohandshakesrdquo that record close contact between mobile phone users In most architectures these proximity data are stored in the usersrsquo phones as anonymized ldquobeaconsrdquo that cannot be used to re-identify the users directly If a user with a PPPT app installed on their phone tests positive and enters test results into their app those who have been identified as having been in close proximity to them can be notified by the app This notification can be automatic or at the discretion of the person who is COV+ depending on the app design If notified a user who has been in contact with a COV+ individual would receive a push notification alert-ing them to possible exposure (which may be timestamped) but with no other identifying information
Because of its reliance on anonymized data PPPT on its own is dis-tinct from manual contact tracing In recognition of this fact some de-signers and researchers now use the more descriptive term ldquoexposure no-tificationrdquo Moreover the public health usefulness of PPPT is uncertain it is unclear how PPPT can best be used in tandem with manual contact tracing especially if the data it collects are inaccessible to or unusable by public health authorities It remains to be seen whether PPPT will provide significant benefit operating alongside but not integrated into manual contact tracing
Between these extremes there are a number of possible middle-ground approaches that aim to strike a balance among public health utility tech-nological feasibility and user privacy protections This middle ground divides into two rough categories centralized storage of de-identified data and decentralized storage of personally identifying data The United Kingdomrsquos NHSX is reportedly developing an app that would utilize BLE handshakes to collect anonymized proximity data which would then be stored on a centralized government-operated server
The most prevalent middle-ground approach in the United States context involves the collection and storage of personal datamdashincluding identifying information and location datamdashon the userrsquos phone This de-centralized but personally identifiable data can then be voluntarily shared with public health officials if the user tests positive for SARS-CoV-2 For
Digital Technology and Contact Tracing 39
example an MIT team has developed an app called Private Kit Safe Paths (MIT nd) that stores usersrsquo location data on their phone for 28 days If a user tests positive she can voluntarily upload her location data to a website that is accessible only to public health officials Officials can then analyze this personally identifiable data and subsequently broadcast re-dacted and de-identified data to other users Healthy users would have access to these redacted location data of COV+ users but their own data would not leave their phones (The developers plan to incorporate BLE proximity data once available) Along similar lines the North Dakota state government has rolled out an app that stores both location data and proximity data on a userrsquos phone which can be voluntarily released by the user to public health authorities if the user tests positive (NDDoH 2020) At a minimum the storage of user location data can function as a ldquomemory aidrdquo if the user tests positive But releasing the data to public health authorities may help them analyze the spread of COVID-19 and alert individuals or groups that have been in contact with persons who are COV+ An overview of various DCTT apps and platforms as well as features that are relevant to this analysis are provided in Table 1
Because DCTTs are so new very little is known about their actual utility to public health authorities for controlling this pandemic Although multiple countries that have had success in greatly reducing transmission of SARS-CoV-2 have included DCTT in their response these countries have employed multiple simultaneous approaches to controlling the vi-rus including manual contact tracing and it is difficult to disentangle what made those responses successful Preliminary impressions from Ice-land may suggest that DCTT at least in that context had a small impact on reducing transmission ldquoespecially compared with methods of manual contact tracing such as phone callsrdquo (Hadavas 2020) This is with the highest public download rate of any DCTT app thus far
DCTTs have the potential to be helpful but they also have the poten-tial to distract from other public health efforts including manual contact tracing Concerns about implementation of DCTT from the public health perspective include that data generated may not be useful to public health authorities either because they donrsquot include detailed data to aid contact tracers or because the data are inaccurate (Mills Rodrigo 2020) DCTT if not calibrated well could be overly inclusive and create many false positives This would be harmful to those individuals being notified and
Purp
ose
Tech
nolo
gies
Use
dD
ata
Stor
age
Part
icip
atio
n
Proximity-based exposure notification
Digital contact tracing (DCT)
Bluetooth LE
GPS
SMS
Centralized
Decentralized
Mandator (actually or functionally)
Voluntaryopt-in
Gov
ernm
ent a
cces
sA
pp N
ame
Dev
elop
er o
r Cou
ntry
Max
WeC
hat
Alip
ayCh
ina
D
ata
com
es fr
om g
over
nmen
t so
urce
s lo
catio
n da
ta s
ent t
o po
lice
Intervention Type
Middle Ground
Trac
e To
geth
erSi
ngap
ore
M
anda
tory
gov
ernm
ent
acce
ss if
pos
itive
NH
SXO
xfor
dO
xfor
d
Gov
ernm
ent m
aint
ains
dat
a
Nex
tTra
ceFr
ed H
utch
inso
n Ca
ncer
Re
sear
ch C
ente
r
Gov
ernm
ent m
aint
ains
dat
a
but n
o st
orag
e
COVI
D
Safe
Path
sM
IT
Volu
ntar
y up
load
by
user
s
who
test
pos
itive
Aar
ogya
Set
uIn
dia
A
nony
miz
ed a
ggre
gate
Care
19N
orth
Dak
ota
In
agg
rega
te o
ptio
nal i
f pos
itive
Minimal
Covi
dSaf
eU
niv
of W
ashi
ngto
n
Non
e
Covi
dWat
chU
niv
of S
tanf
ord
amp U
niv
of W
ater
loo
To
val
idat
e te
st re
sults
CoEp
iCo
Epi
O
pt-in
to s
hare
BT
and
sym
ptom
lo
g w
ith C
oEpi
ser
ver
itoG
erm
any
N
one
pos
itive
resu
lts to
ito
serv
er
TAB
LE 1
Ex
ampl
es o
f Dig
ital C
onta
ct T
raci
ng T
echn
olog
ies
to S
uppo
rt A
ctiv
e Pu
blic
Hea
lth S
urve
illan
ce a
nd R
elev
ant F
eatu
res
Digital Technology and Contact Tracing 41
asked to quarantine unnecessarily and it could result in large proportions of the population remaining at home at any one time Individuals living or working in congregate settings could receive frequent notifications that would result in their inability to leave quarantine for long periods of time Finally public health authorities could also become inundated by data from these technologies and not have sufficient approaches to manage or analyze the incoming information
Relevant Differences between Manual and Digital Contact Tracing
There are several noteworthy differences between manual contact tracing efforts and use of DCTT First there is a significant amount of evidence regarding the effectiveness of manual contact tracing which is lacking for DCTT Second manual contact tracers interact with individuals who are confirmed or suspected cases and contacts of cases but not other members of the general public DCTT intervention would affect all users regardless of circumstances (though some more than others) Third manual contact tracing occurs most often through human-to-human encounters with the opportunity to clarify misconceptions address worries and express sympathy and other important affects DCTT can certainly incorporate sharing of important information and potentially communicate some af-fect but it currently lacks a range of other human capabilities and char-acteristics Fourth there typically are fewer data intermediaries in manual contact tracing (fewer entities handling data) in DCTT a valid argument could be made that a wide range of technology developers (and perhaps mobile network operators) must remain connected to relevant data in order to continuously identify problems and improve functionality
It is because of these and other differences that DCTT has been pro-posed as a potential complement to rather than a replacement for man-ual contact tracing However over time it is possible that technology could develop to close gaps between some of these differences (if and as needed) and in parallel the goals of contact tracing and public health surveillance may evolve
Ethics of Designing and Using DCTT
43
THREE
43
Those developing DCTT and those considering its use should systemat-ically take into account and document alignment with the guiding princi-ples outlined in this report
When considering the ethics of DCTT key ethical questions con-cern the features that DCTT should have (eg should digital contact tracing apps collect usersrsquo location data) whether and how individualsrsquo data should be shared with public health authorities how ethically to encourage use of DCTT (eg under what circumstances would it be eth-ical to incentivize or mandate use of DCTT) what kind of supports and equity-promoting measures should accompany use of DCTT and how governance and oversight of DCTT should be structured
The sections that follow consider these questions one by one A key conclusion of this report is that these features of the design and use of DCTT are ethically interrelatedmdashreaching a determination regarding any one question requires careful consideration of them all Rather than reaching ldquoone size fits allrdquo conclusions about specific features of uses of DCTT decision makers should ethically assess DCTT systems holistically
Generally a public health measure is ethically justifiable if it strikes a reasonable balance between competing considerations and if it pro-vides sufficient public health benefit (or the prospect of benefit) to justify the burdens associated with it DCTT systems are ethically justifiable if they strike a reasonable balance between multiple ethical considerations including
bull enabling an effective and efficient public health response
bull protecting individual privacy and preventing harms to individuals
44 Digital Contact Tracing for Pandemic Response
including harms from sensitive data being revealed and from erro-neously being subjected to isolation or quarantine orders
bull allowing individuals to control what information about them is collected and revealed to whom including through appropriate dis-closure and authorization processes for data collection
bull promoting equitable distribution of benefits and burdens of DCTT
bull maintaining public trust in DCTT and in the COVID-19 public health response and
bull taking seriously the future implications of decisions that we make today
To illustrate a holistic assessment consider whether it is ethically jus-tifiable for an employer to mandate that employees use a DCTT as a condition of returning to work This will depend upon many features of the DCTT system what kind of data the DCTT collects (eg does it collect location data or just record proximity events) whether there is public health capacity to make good use of these data what the data are used for (eg will the employer ban an employee from the workplace on the basis of a DCTT-identified contact) what kind of social supports are available (eg is there paid leave for employees) what employeesrsquo attitudes are toward use of DCTT and whether mandating use is likely to have public health benefit among other factors These factors may vary from place to place and may change over the course of the pandemic Thus there is no ldquoone size fits allrdquo ethically optimal approach to DCTT
Justifying the Use of DCTT Systems
A foundational issue is why deploying any DCTT during a pandemic is justified given there are manual contact tracing capabilities that are well established while the performance and effectiveness of novel technolo-gies is less established The need to move quickly to minimize the spread of the virus poses challenges here as the data needed to fully make the case that these technologies substantially contribute to the public health response may not be available prior to widespread use The primary ar-gument for DCTT is that the capacity of manual contact tracing may be
Ethics of Designing and Using DCTT 45
exceeded and we may not be able to bolster the public health workforce rapidly and sufficiently enough to meet needs DCTT has the potential to quickly and exponentially expand the reach of contact tracing In ad-dition DCTT may allow more efficient identification and quarantine of potential contacts of COV+ people than manual contact tracing alone particularly given the high number of infections that have been spread by asymptomatic individuals
Nonetheless reasonable people disagree about the prudence of pur-suing DCTT especially given its limited performance history and poten-tial risks including diverting attention and resources from more effective interventions The limited attention and resources available during a pan-demic must be allocated efficiently and effectively
To justify potentially widespread use of technologies such as DCTT therefore a number of considerations must be addressed
bull whether the technology is designed to meet an important and unmet public health need
bull whether there is sufficient evidence or reason to suggest that the technology will be effective at serving its purpose
bull whether the outbreak is characterized by sufficiently severe morbid-ity and mortality and a high rate of disease transmission to warrant large-scale introduction of novel systems
bull whether there are other less autonomy-restricting or less risky al-ternatives to widespread use and
bull whether it is reasonably likely that a sufficient number of individ-uals will use the technology to achieve the intended public health benefit
Monitoring and Evaluating Technologies to Inform Policy and Practice
A number of public health ethics principles necessitate the ongoing mon-itoring and evaluation of DCTT systems First DCTT must be shown to perform reasonably well at achieving its stated goal reducing the spread of SARS-CoV-2 The effectiveness of DCTT programs should be illus-trated at a number of stages
46 Digital Contact Tracing for Pandemic Response
1 Robust initial technology testing is needed to publicly justify the widespread adoption of DCTT and avoid public failures which may hamper future uptake (eg Lovejoy 2020 Morse 2020) This typically includes alpha testing in virtual environments and beta testing in different community settings
2 If and when a DCTT is implemented on a wide scale it must be monitored on an ongoing basis to assess reach effectiveness func-tionality best practices and any harms
3 When approaching a previously identified stopping point for use of DCTT monitoring can help to identify when utilization is no longer needed
If at any of these points evidence clearly suggests harm (particularly in comparison to other methods that the public might find more acceptable) this evidence should provide a basis upon which to revisit strategies pri-orities and allocation of resources Attention should be given to foresee-able side effects that may dramatically influence the overall effectiveness of the program such as individuals carrying their smartphones around with them selectively so as to avoid particular undesired consequences of DCTT policies
Anonymized aggregate data including user feedback must be eval-uated to ensure that benefits and burdens are distributed fairly As noted earlier unintended burdens may include inequitable outcomes that may arise in a DCTT program for example resulting from uneven access to the required technology to participate disparate concerns about sur-veillance within some communities that might limit widespread use or discrimination that may result from being identified as COV+ due to the program or for communities that are termed ldquohotspotsrdquo based on maps of COV+ location data Additionally it is possible that some communi-ties might get higher rates of false positives because they are located in densely populated areas thus increasing the burden of self-isolation If any of these inequities are identified steps must be taken to mitigate them
Finally numerous actors should engage in the monitoring and eval-uation of DCTT systems Technology developers and public health re-searchers have a clear role in this process Technology developers should work with public health researchers to monitor accuracy precision func-
Ethics of Designing and Using DCTT 47
tionality confidence of estimates sources of error and the like Research-ers may also be able to contribute innovative methods to systematically and rapidly evaluate candidate technologies such as by deploying cluster randomized stepped wedge (Hemming et al 2015) or adaptive trial de-signs and techniques (eg response-adjusted randomization) (Pallmann et al 2018) These approaches were also proposed for use in research to assign candidate experimental treatments and vaccines during the 2014-15 Ebola outbreak (Berry et al 2016) When formal research activities are pursued ethics principles and legal requirements for the conduct of research should apply (eg The Belmont Report)
Furthermore any workplace or institution that incentivizes or man-dates use of DCTT has a responsibility to provide evidence that the in-tervention at minimum is not likely to cause harm and to monitor for unanticipated burdens In all cases it is vital that a trusted intermediary be involved in the evaluation of DCTT programs to limit perceptions of bias and ensure a legitimate basis for decision-making Nonsensitive aggregate DCTT analyses should be made available to the public so as to permit verification and inform continuing public debates about its useful-ness and necessity At an individual level data should be available to users that would permit them to further investigate their personal risk with public health officials or other health workers This is important not only to ensure their health and well-being but also to add a layer of protection against unnecessary quarantine
Recommendations
bull Reviews of DCTT systems must be conducted in part by an inde-pendent intermediary that has established the publicrsquos trust
bull Those who authorize use of DCTT within a particular jurisdic-tion or institution should continuously and systematically monitor the technologyrsquos performance in that context This should include monitoring for effectiveness and benefit monitoring for harms and monitoring for the fair distribution of both benefits and harms They should also monitor evidence that is being generated in other contexts about their selected technological solution and about other competing technologies
48 Digital Contact Tracing for Pandemic Response
bull Data should be available to users that would permit them to further investigate their personal risk with public health officials or other health workers to add a layer of protection against unnecessary quarantine
Public Trust and Public Attitudes
Researchers have estimated perhaps conservatively that DCTT use by 80 of smartphone ownersmdash56 of the populationmdashwill be needed to suppress the epidemic (Hinch et al 2020) These estimates also highlight that some decrease in transmission would be realizable even with lower rates of technology adoption As such in order to maximize impact it is essential to gain a thorough understanding of public perspectives on DCTT including which features and uses of the technology the pub-lic finds acceptable which kinds of DCTT the public would be most likely to use and which designs and uses of DCTT would maintain or jeopardize public confidence and trust There will be variation in public attitudes within and across societies and over time
With respect to what we currently know about public attitudes and trust in DCTT in the United States polling data suggest some potential support and also some divisions regarding willingness to use the technol-ogy Polls conducted by groups based at the University of Zurich (Hargit-tai et al 2020) and the University of Oxford (Altmann et al 2020) suggest that more than 60 of Americans would be willing to install such an app Both a Washington PostndashUniversity of Maryland poll (2020) and a Kaiser Family Foundation poll (Kirzinger et al 2020) show roughly half of the population would be willing to install the app Over half of the population (59) would be willing to share their COVID-19 positive test result with an app in order to anonymously share that information with their contacts (Washington PostndashUMD 2020) Only 29 of respondents to a March 12ndash27 Oliver Wyman Forum poll (Elliott et al 2020) said that they would be willing to share their location data Additionally Washing-ton PostndashUMD data and Pew data from 2019 suggest that approximately one in six Americans do not have a smartphone and thus cannot use the technology without intervention (Pew Research Center 2020)
People may be more willing however to download an app if it will
Ethics of Designing and Using DCTT 49
ease social distancing policies and allow for more economic and social activity Willingness to install a contact tracing app increased among re-spondents to the Kaiser Family Foundation poll from 50 to 66 when respondents were asked if they would be willing to do so to allow schools and businesses to reopen Additionally who develops or administers the app appears to matter Respondents to the Washington PostndashUMD poll indicated higher levels of trust that their anonymity would be preserved by public health agencies and universities than by tech companies or health insurance companies Further more respondents to the Oliver Wyman Forum poll were willing to share their health information with public health authorities (55) than the local government (35) their employer or school (33) or the federal government (27)
These data suggest that people will be more willing to use a contact tracing app when the potential benefits are clearly identified and valued such as lifting social distancing measures and they will be more willing to do so if the data are going to a public health agency rather than the federal government or a tech company Other factors that seem to be as-sociated with greater willingness to install a contact tracing app include younger age and the app source (Hargittai et al 2020) with a preference for apps distributed by public health agencies over others such as health insurers or public universities (Hargittai and Redmiles 2020) However all of this must be read with caution as public polling may not be repre-sentative of some populations or of widespread public attitudes Further these attitudes may shift over time and may be discordant with behaviors (Barth and de Jong 2017)
Deliberative public engagement efforts would be an appropriate means of filling in gaps in understanding about the acceptability of dif-ferent approaches (Fishkin and Laslett 2003 Cavalier 2011) In addi-tion including the public particularly in the earlier stages of planning a path to sustainable resolution to the pandemic could serve to help disseminate a nuanced understanding of what is at stake including the key challenges and trade-offs Aggregated public polling results are not sufficient as a proxy for careful analyses of the ethical challenges but they do provide a necessary input for these analyses Integrating lessons and outputs from public engagement into guidance and other products requires special attention and should be validated and enhanced through further engagement
50 Digital Contact Tracing for Pandemic Response
Recommendations
bull More research into public attitudes is needed In particular in-depth qualitative research should examine public attitudes about percep-tions of trust in DCTT among different communities which fea-tures of DCTT influence trust and the extent to which people are willing to provide different types of data through DCTT to help their community
bull States and localities that are considering adopting DCTT should engage with the public to increase their understanding of the ac-ceptability of DCTT design features and uses among diverse communities
Designing Flexible Technology to Maximize Public Health Utility While Respecting Other Values
Values in Design
Efforts to advance DCTT in the United States and elsewhere have empha-sized the importance of ldquoprivacy by designrdquo that is building privacy and security protections into the design of technology rather than counting on responsible use alone (Cavoukian 2010) As noted above some major technology companies have signaled this position through development of decentralized privacy-preserving proximity tracking (PPPT) systems These systems embed features such as decentralization anonymity of us-ers bans on collection of location data and minimal reliance on or inte-gration of public health authorities or other government actors Many of these features have also been embraced early by advocacy organizations (Crocker Opsahl and Cyphers 2020 Electronic Privacy Information Center 2020 Kahn Gilmor 2020) and in an open letter from nearly 300 researchers (ldquoJoint Statement on Contact Tracingrdquo 2020)
Privacy by design provides principles that incorporate one set of val-ues (privacy) into the design of DCTT Importantly the principles ac-knowledge the need to design privacy defaults into systems while main-taining the capacity of those systems to achieve their otherwise justifiable ends Put another way privacy by design ldquoembraces legitimate non-pri-vacy objectives and accommodates them in an innovative positive-sum mannerrdquo (Cavoukian 2010 p 4)
Ethics of Designing and Using DCTT 51
This stance simple in its statement is not easy to satisfy Given that ldquoobjectivesrdquo are themselves driven by values it begs for an articulation of additional values (aside from privacy) that individuals and groups within societymdashincluding many privacy advocatesmdashmay believe to be important For example at any moment in addition to valuing their own privacy individuals may value efficiency equity autonomy economic well-being companionship patriotism or solidarity Moreover the above stance necessitates an acknowledgment that peoplesrsquo value priorities often change when circumstances change not least of which during a pandemic when mass physical distancing has made it difficult to fully realize many important values (aside from physical privacy) A different orientation is needed at this moment As Flanagan Howe and Nissenbaum (2008) conceptualized in 2008 we should take a ldquovalues in designrdquo approach to DCTTmdashan approach that designs a broader range of values such as those enumerated above into technology
This approach requires a wider ethical lens through which to ex-amine DCTT and requires hard but important work to appropriately balance competing interests within technology architecture For example there is value in technology providing users the option to collect their location history and share it with public health professionals in order to advance the public health response increase system efficiencies (eg by contributing information that can lead to better data processing) and reduce the burden on essential workers For some this might be an ex-pression of autonomy solidarity or patriotism At the same time there is value in further advancing autonomy by designing technology to allow individuals to control what data about them are collected and shared
Justifying a Middle-Ground Approach to DCTT
We ought to embrace a DCTT that has a default of interoperability and privacy protection but that does not stop there Triggering events such as entry of a positive test result or receipt of a notification that one was proximate to someone who tested positive could for example generate a push notification that users can acknowledge in order to permit transmis-sion of potentially useful location data to public health authorities This could be accompanied by an explanation of the value of the information and relevant restrictions on its use
At this point it is worth reiterating that manual contact tracingmdash
52 Digital Contact Tracing for Pandemic Response
which involves collecting information from people whorsquove tested posi-tive and their contactsmdashincludes collection of personal information and potentially embarrassing or sensitive data about the places theyrsquove been and the people theyrsquove had contact with Manual contact tracing efforts use these data to uncover ongoing transmission provide useful informa-tion tailored to the individual and enable isolation and quarantine as necessary
It stands to reason that if these forms of data can be collected by a DCTT and provided to public health authorities in a maximally secure and voluntary way (with clear rules regarding authorized uses) this may amplify public health authoritiesrsquo manual contact tracing efforts For ex-ample location data from DCTT could help jog peoplersquos memories about where theyrsquove been and fill in memory gaps This is especially relevant given the long period of infectivity of SARS-CoV-2 which begins before people are symptomatic and therefore before they are aware they are infected (Ferretti et al 2020) Location data might reveal that a COV+ person was at a restaurant at an exact time and date which could be followed up by contact tracers who could alert the public or use other measures to reach those who were also present in the restaurant at the same time In other disease contexts (see Furlanello et al 2002 Dredze et al 2013 Eckhoff and Tatem 2015 Fraccaro et al 2019) geolocation data have demonstrated some potential to support epidemiology and dis-ease surveillance with technical cautions regarding accuracy and the like (Beukenhorst et al 2017)
These benefits are currently speculative for DCTT At present pro-viding public health authorities with large amounts of data on cases and potential case contacts will be useful only if there is sufficient public health capacity to follow up on these data In addition there is a risk of low-quality data from DCTT flooding the system Investigating poten-tial case contacts identified by a DCTT may distract them from other important efforts and at some point overwhelm public health capacity altogether Whether and to what extent data from DCTT will benefit contact tracing efforts is unknown pointing again to the importance of continuously collecting high-quality evidence about DCTT
Nevertheless what would enable the most flexible and potentially robust public health response is to design DCTT so that restricted data sharing is possible From an ethics perspective the collection and use
Ethics of Designing and Using DCTT 53
of sensitive data in manual contact tracing efforts (described above) is typically seen as ethically justifiable so long as there is sufficient public health benefit and need Thus wouldnrsquot it seem appropriate from both a public health and ethics perspective to design DCTT systems to enable similar data to be shared with public health authorities when and if there is ethical justification for sharing them
Why instead do so many advocate that DCTT should be designed as a ldquominimalrdquo system when this arguably ties the hands of public health and individual users and precludes the collection of data that public health authorities (and indeed many other apps on our phones) typically collect We here consider and appraise some of the reasons that may motivate individuals and groups to argue for minimalistic positions
1 Proponents of minimal systems may believe that such systems will be most
widely adopted Some groups have maintained that only these systems will earn and maintain public trust and be widely adopted (Simpson and Conner 2020) For example the previously referenced open letter (ldquoJoint Statement on Contact Tracingrdquo 2020) asserts ldquoSome of the Bluetooth-based proposals respect the individualrsquos right to privacy whilst others would enable (via mission creep) a form of government or private sector surveillance that would catastrophically hamper trust in and acceptance of such an application by society at large It is crucial that citizens trust the applications in order to produce sufficient uptake to make a difference in tackling the crisis It is vital that in coming out of the current crisis we do not create a tool that enables large scale data collection on the population either now or at a later time Thus solutions which allow reconstructing invasive information about the population should be rejected without further discussionrdquo
Response While it is true that public trust in and acceptance of DCTT is essential for its success there is insufficient evidence that public trust would be threatened by a DCTT system that has the capacity to collect location data and enable voluntary sharing of those data with public health authorities A contrasting perspec-tive is that maintaining public trust requires maintaining public confidence that the DCTT system is providing useful information is benefiting and not harming individuals and is advancing the
54 Digital Contact Tracing for Pandemic Response
public health response (Leprince-Ringuet 2020) From this per-spective a system that is less well integrated into the broader public health response or that generates a higher rate of false positives (as some suggest decentralized approaches might (Fraser et al 2020)) may fare worse when it comes to maintaining public confidence and trust
2 Proponents may hold the view that minimal systems are harmless (or nearly
harmless) to individuals This is because individuals are anonymous none of their location data are gathered and none of their identifiable data are shared with anyone In contrast DCTT systems that collect and share identifiable data including location data may be seen as posing risks of harm to individuals
Response While minimal systems may be harmless (or nearly harm-less) from the perspective of protecting privacy they may not be harmless from the perspective of public health if they generate system inefficiencies through producing too many false positive or false negative contacts Aside from presenting a challenge for public health professionals false positives could also harm individ-uals If users receive a large volume of automated messages alert-ing them to proximity events will this cause distress Will a large volume of alerts cause users to become disengaged and stop using the DCTT or lose confidence in contact tracing more generally as a legitimate method of disease control Admittedly these are just potential harms and risks it is unknown the degree to which they will materialize The point is that privacy-related harms are not the only relevant harms to individuals that we should consider when assessing DCTT
We acknowledge the risk under a middle-ground DCTT of data being used in ethically unjustifiable and harmful ways For exam-ple it would be against the principles and recommendations artic-ulated in this report for data to be sold or monetized by technology companies or others for corporate gain and this misuse of data would be more intrusive if the data were potentially identifiable What makes it ethically justifiable to take this risk is the compen-sating benefit of allowing the most flexible and robust public health
Ethics of Designing and Using DCTT 55
response during the pandemic but this alone is not sufficient The risk of inappropriate uses must be reduced by ensuring stringent requirements for data security and access as well as clear legal protections and recourse for any violations (as discussed further below)
3 Proponents may believe that DCTT systems should not collect location data
as this would be too intrusive and of insufficient value Some proponents of PPPT systems maintain that recording proximity events is sufficient and data relating to usersrsquo movement and location should not be collected (Ingram 2020) The thought may be all we need to know is whether two individuals came into close enough contact for viral transmission to have occurred we donrsquot need to know where or when this contact occurred and there is no need to collect and store usersrsquo location data
Response This conclusion might be too hasty As discussed above there is potential (though unproven) benefit to providing public health authorities with location data Location data could help jog peoplersquos memories about where theyrsquove been provide more context for understanding the nature of ldquoproximity eventsrdquo captured by the DCTT and allow public health authorities to quickly define a category of individuals who may be at risk Collecting location data from cases is what public health authorities do on a regular basis following best practices for manual contact tracing
In addition many peoplersquos location data are currently gathered by apps on their phones and used for various purposes such as to provide more accurate navigation to offer entertainment or to improve services Many are willing to accept these capabilities because they provide some value in return Why not allow DCTT to also collect these data so that the data are available for users to share with public health officials who can then do their work more effectively and refine their understanding of how the disease transmits If many are willing to have these data used to find a bet-ter route home why not let individuals share these data to support the effort to save lives
56 Digital Contact Tracing for Pandemic Response
4 Proponents may hold the view that minimal systems pose little or no threat
to individual autonomy whereas systems that collect identifiable data and
integrate public health do pose a threat to individual autonomy For exam-ple they may worry that use of DCTT could be mandated and not a voluntary choice and in this circumstance mandatory use of minimal DCTT would be less intrusive risky and privacy violating Another worry might be that itrsquos theoretically possible that DCTT could share individualsrsquo data with public health authorities without usersrsquo full understanding if the technology does not even gather identifiable data then itrsquos not possible for these data to be shared without the individualrsquos consent
Response We discuss the importance of appropriately designed disclosures and consent below as well as the high bar that would need to be met to ethically justify mandatory use At this time mandated use of DCTT by states or institutions is not justifiable given uncertainty about potential harms and benefits Users should have a meaningful opportunity to review and understand infor-mation about the specific technology and its uses and to consent Assuming that individuals are not required to use DCTT and that they provide consent to using it designing DCTT to make data collection and sharing possible is the design choice that maximizes individual autonomy because it provides individuals with options they may value
Individuals may wish to share their data with public health au-thorities for both self-interested and altruistic reasons For exam-ple someone who has tested positive for SARS-CoV-2 and enters this test result into an app may wish to be connected to public health authorities in order to be provided with needed information resources and support She may wish for public health authorities to be provided with her phone number in case they need to reach her to provide additional information Further someone who has been alerted by an app that he had a ldquoproximity eventrdquo with a person who has tested positive for SARS-CoV-2 may wish he had location data to share with public health authorities in order to help ascertain whether this event is a cause for concern or whether it is likely a false positive (eg he and the COV+ person were sepa-
Ethics of Designing and Using DCTT 57
rated by a wall) Someone who tests positive for the virus may also wish to share their location history with public health authorities in order to be as helpful as possible to the overall public health re-sponse by facilitating de-identified aggregate analyses that identify locations of higher transmission or contribute to refining overall understanding of the disease and pandemic
5 Concerns about ldquosurveillance creeprdquo and the long-term downstream effects of
digital contact tracing system may also motivate embrace of minimal DCTT Digital contact tracing technology that collects identifiers and loca-tion data and has the capacity to share them with public health au-thorities may represent a massive and concerning increase in govern-ment surveillance of the public It might be feared that the use of this surveillance capacity in the COVID-19 response sets an unwelcome precedent for future use in other contexts Designing DCTT as min-imal systems may be a way to minimize the risk of surveillance creep and to minimize the harms associated with potential future uses of the technology
Response Surveillance creep is a serious concern To guard against surveillance creep protections should be put in place to ensure that only those data that are necessary and relevant for the public health purposes at hand are collected and used and data should be kept only for the period of time needed for those public health purposes In the face of these concerns it is important to emphasize that widespread use of DCTT in the COVID-19 response is justi-fied by the exceptional circumstances of the current pandemic and their use in this context does not imply that future public health use is ethically appropriate without significant public debate (eg use in seasonal flu surveillance efforts) Future use will require in-dependent justification Use of DCTT in other contexts (eg law enforcement or immigration enforcement) is also presumptively unethical
All in all the arguments that DCTT should be designed as a minimal system are not convincing Rather DCTT should be developed through a ldquovalues in designrdquo approach with a core set of features that protect pri-
58 Digital Contact Tracing for Pandemic Response
vacy with enough flexibility to be used differently depending upon local conditions evolving evidence and individual preferences What kind of digital contact tracing system will strike the right balance between public health goals and other considerations will depend upon circumstances For example whether it is even beneficial to provide public health au-thorities with volumes of data about potential contacts of COV+ people will depend in part upon whether they have the capacity to make good use of those data This will vary from location to location and will change over time
Recommendations
bull Technology companies should not alone control the terms condi-tions or capabilities of DCTT nor should they presume to know what may be acceptable to members of the public
bull A ldquovalues in designrdquo approach to development of DCTT should be adopted (Flanagan Howe and Nissenbaum 2008 Knobel and Bowker 2011) Robust public- and user-engagement activities should be pursued to identify and incorporate to the extent pos-sible a range of values into the design of the technology These values may include privacy but also autonomy efficiency equity or others Technology design should reflect an appropriate balance and prioritization of identified values
bull Technology design should not be static but rather it should be ca-pable of evolving depending upon local conditions new evidence and changing preferences and priorities
bull DCTT should be designed to have a base set of features that protect privacy with layers of additional capabilities that users may choose to activate An initial default should be that user location data are not shared but users should be provided with easy mechanisms and prompts to allow for opting-in to this capability with encourage-ment to the public if and as it is shown to be critical to achieving public health goals
Ethics of Designing and Using DCTT 59
Policy Positions to Advance Widespread Use of Digital Contact Tracing Technologies
The public health value of a DCTT depends in part on the number of people who use it This section concerns broad public policy positions that relate to the widespread adoption of DCTT What are ethical means of encouraging or securing widespread adoption of DCTT systems Un-der what circumstances would it be ethical to mandate their use or incen-tivize their use What enforcement challenges exist
Mandating Use
Digital contact tracing has occurred without the publicrsquos explicit volun-tary agreement in some countries such as China and Israel In others use has been voluntary (Valentino-DeVries Singer and Krolik 2020) For example Singapore adopted an app that the public could use on a vol-untary basis and approximately 20 of the population has downloaded and used it Norway has recently launched a contact tracing app that was downloaded by roughly 30 of the population in the first week that it was made available In the United States many advocates and researchers have argued that use of digital contact tracing tools must be fully volun-tary this is the dominant perspective
There are numerous ways that DCTT could be put into use without user choice For example as has been done in Israel location data from mobile phones could be collected and used by the government without usersrsquo consent Use of an app could be formally mandated as a precon-dition for returning to work or school or even further to control entry into a facility or onto transportation such as airplanes through scanning of a QR code to demonstrate personal exposure levels (Gan and Culver 2020)
While these approaches are hard to imagine in the United States some contend that mandatory use of digital contact tracing tools could be ethical and may even be ethically required Mandating use of digital contact tracing tools could in theory vastly increase the effectiveness of digital contact tracing systems and thus may save more lives and allow states to lift lockdowns sooner or avoid reimposing lockdowns in the future Canca (2020) argues that use of privacy-by-design digital contact
60 Digital Contact Tracing for Pandemic Response
tracing tools should be mandatory because the use of these tools will be nearly harmless if there are sufficient privacy protections In addition mandatory use of DCTT that embraces these principles is significantly less intrusive at the individual level than manual contact tracing which involves the collection of personally identifying and potentially sensitive data In this light it could be argued that such mandates are actually pref-erable from the perspective of both public health and individual liberty insofar as they reduce the likelihood of ldquostay at homerdquo orders which are a severe limitation of individual liberty
Nevertheless mandated use of DCTT systems faces considerable ob-stacles For example people may not adhere to the mandate by simply leaving their phone at home thus preventing their activities from being tracked Even more harmful would be if people react to a mandate and a perceived violation of liberty and privacy by employing location and Bluetooth spoofing software to shield their real contacts behind a screen of misinformation The introduction of this misinformation into a contact tracing effort might severely undermine its effectiveness The possibility of nonadherence also raises the issue of enforcement would high rates of nonadherence be permitted or would enforcement be attempted (if even possible) Perhaps more important should the technology not deliver the hoped-for benefits having mandated the use of an unproven technology could result in a loss of public trust in the technology the entity instituting the mandate and potentially the larger public health response (Bernstein et al 2019)
Mandatory DCTT could also be used to enforce quarantine restric-tions and stay-at-home orders for those who are COV+ or are determined to be at heightened risk The use of DCTT in enforcement activities raises a number of ethical (and legal) issues that are beyond the scope of the present analysis In particular individuals have a heightened interest in personal privacy if their data can be used to restrict their freedom of movement and other civil liberties At a minimum stringent procedural protections would be required to ensure that the data collection is fair and unbiased and that DCTT users are provided with adequate informa-tion in advance about how their data may be used
Mandatory use policies for DCTT must therefore convincingly ad-dress a number of questions including
Ethics of Designing and Using DCTT 61
bull Is the technology designed to meet an important and unmet public health need
bull Is there sufficient evidence to suggest that the technology will be effective at serving its purpose
bull Is the outbreak characterized by sufficiently severe morbidity and mortality and a high rate of disease transmission
bull Are there other less autonomy-restricting or less risky alternatives to widespread mandatory use of DCTT
bull Is it possible and likely that a sufficient number of individuals will comply with a mandate
bull Can inequities in the burdens and benefits of the mandate be suffi-ciently addressed through social protections and countermeasures
bull Can enforcement and enforcement discretion be implemented in a manner that is consonant with fundamental rights
bull Will those subject to the mandate interact closely with a population that is at high risk of morbidity or mortality if they contract the virus
bull Is it possible to mandate use and remain consistent with important ethical and legal principles
These questions would need to be satisfactorily addressed and explicitly documented by any decision maker considering mandatory use includ-ing government officials institutional leaders and employers Particularly important is the need to identify reliable evidence that the DCTT would be effective and to ensure that the burdens and benefits of use are equita-ble and justifiable If use of a DCTT is a condition for returning to work or school in person those who refuse or are unable to use DCTT should not lose their jobs or positions as a result and adequate support should be in place for people who are asked to self-quarantine
Finally it is important to distinguish a mandate from a ldquopushedrdquo program installation or a default setting in an application which can be modified by users A mandate relates to a policy of required use whereas the pushed programs or default settings relate to the chosen architecture for download and operation of the application
62 Digital Contact Tracing for Pandemic Response
Incentivizing Use
Perhaps the most effective way to generate widespread adoption of DCTT in the United States is to offer incentives to individuals who choose to adopt and who properly utilize the preferred DCTT approach in a voluntary system External incentives may help ldquonudgerdquo populations toward desired adoption targets Given the importance of widespread use of DCTT modest incentives ought to be considered for DCTT in the US if and when there is sufficient evidence of the technologyrsquos utility Note that in other contexts studies have shown that the provision of some incentive leads to an increase in adoption or utilization of public health programs (Singer and Ye 2013 Lee et al 2014) Moreover even a relatively small incentive can achieve much greater rates of adoption with some studies demonstrating that the incremental adoption gain de-creases as the incentive gets larger (Thornton 2008 Gibson et al 2019) In the context of COVID-19 incentives that might be both effective and ethically acceptable could include a relatively small monetary token free or discounted mobile phone service for a period of time or credit to be used by means of a mobile phone
Not all incentives are ethically appropriate For example making access to lifesaving health care contingent on using a DCTT or making valuable disease information available only to DCTT users would not be ethically appropriate In addition incentives cannot be used to over-come otherwise ethically unjustifiable technology design for example they should not be used as an offset for providing personally identifiable health information to other users
Importantly incentivization schemes must be kept distinct from man-dates as the latter require greater ethical justification To offer an incen-tive is to offer something of actual value to individual participants over and above what they are reasonably entitled to at baseline For example making a return to work contingent on using DCTT is not offering an incentive but instead imposing a mandate and it would have to be justi-fied as a mandate
In the context of COVID-19 it is also necessary to recognize that there is an inherent ldquoincentiverdquo behind the technologymdashthat is the prom-ise of more lives saved faster pandemic recovery and the reduction or elimination of blanket physical distancing Effective public communica-
Ethics of Designing and Using DCTT 63
tion of these goals if and when there is sufficient confidence in the tech-nology is important
Encouraging Use
Another important approach to increasing use of DCTT in the United States is for trusted leaders to encourage their use Community leaders public figures health care professionals and other respected individuals who have the publicrsquos trust and goodwill could be enlisted to commu-nicate with the public about DCTT and encourage its use drawing on notions such as communal responsibility solidarity and so on These en-couragements could be combined with other approaches (eg small in-centives) to optimize reach while continuing to respect individual choice
Recommendations
bull DCTT use should not be mandated at this time given uncertainty about potential harms and benefits Additional technology user and real-world testing is needed
bull Incentives can be a useful complement to encouragements how-ever any incentives for users to install and use DCTT must be equitable should not be coercive and should align with effective use of the technology (ie they should not incentivize downloading an app but then leaving onersquos phone at home)
bull Trusted leaders should be enlisted to communicate effectively with the public about DCTT and encourage its use should the technol-ogy demonstrate some potential The limits of knowledge regard-ing effectiveness should also be explained along with what will be done to improve technological capabilities as understanding evolves
Disclosure and AuthorizationConsent
In deciding whether to use DCTT voluntarily individuals must be suf-ficiently informed both through broad coordinated public engagement campaigns and individual-level disclosures and there must be a meaning-ful mechanism for users to consent It is important to recognize that while
64 Digital Contact Tracing for Pandemic Response
informed consentmdashwhich is characterized by detailed consent forms and requires a witnessed signaturemdashis the standard for most research and clinical care encounters (Faden and Beauchamp 1986) it is not typically the standard for public health disease surveillance In the public health context other relevant protections (such ethics training for public health professionals and strict data handling and confidentiality requirements) are in place and there is a strong public health interest in collecting the relevant data A more limited role for consent has been recommended for public health surveillance based on a reciprocal obligation of members of society to contribute to a ldquocommon goodrdquo and particularly in the con-text of a pandemic practical considerations such as time constraints and exigencies such as increasing morbidity and mortality (WHO 2017)
Under current circumstances given that (1) many individuals have time and capacity to consent (2) DCTT is being considered as part of plans for longer-term restabilization (3) DCTT is not a familiar part of our public lexicon (4) remote consent disclosure and authorization can be easily embedded in DCTT systems (Moore et al 2017) and (5) there are justifiable public deficits in trust with respect to various government and corporate actors handling potentially personal digital information a strong ethical case can be made for requiring a carefully crafted version of what is sometimes referred to as simple consent Simple consent consists of basic disclosure and voluntary agreement or authorization (Ali et al 2017) Three questions then arise
1 What information should be disclosed to potential users of DCTT
bull Information disclosed might include
deg Entity responsible for the technology
deg Its purpose
deg How it works (in lay terms)
Some participatory disease surveillance systems (eg Flu Near You) have received for-
mal ldquowaiversrdquo of consent requirements from institutional review boards (IRBs) in the US
As they undergo development these digital surveillance systems often straddle a line be-
tween public health surveillance and research hence the frequent need or desire to obtain
ethical review by an IRB (Ali et al 2019)
Ethics of Designing and Using DCTT 65
deg What users need to do
deg Any user options eg
Sharing geolocation data with public health authorities when that would facilitate a defined public health goal
Sharing de-identified metadata with technology develop-ers (for system enhancement)
deg User rights
deg How data will be handled
What data are collected
What data are shared (and how and with whom)
Purposes for which data can be used and not used
How data are secured and protected
Whether and what data will be retained (or will be deletable)
deg Potential benefits and any known risks
deg How to obtain answers to questions about the technology and public health response
2 How should this information be presented
Information should be presented leveraging eConsent models that are more accessible than long ldquoclickwraprdquo disclosures typical of mobile apps (Iwaya et al 2019) For example a simple open-source smartphone con-sent module that has been developed by Sage Bionetworks for research uses could be adapted to the public health surveillance context and to DCTT (Doerr Suver and Wilbanks 2016)
bull Formatting recommendations include (cf Doerr et al 2016)
deg simple and straightforward information
deg deliberately organized content
deg multimodal learning (eg visual audio written)
deg accessibility for disabled users
66 Digital Contact Tracing for Pandemic Response
deg multilingual text
deg engagement through interaction (eg swiping to navigate forward and backward)
bull The same simple information should be made publicly available via multiple other platforms (eg on websites in newspapers over social media)
bull More detailed disclosures should be made readily accessible to those who wish to learn more with no hidden surprises
3 How should users signal that they agree to the details specified in disclosures
Opt-in Models
Opt-in models are those that through an affirmative act such as clicking a button users would indicate their intention to use a DCTT This ap-proach is consistent with other app downloads where app details and privacy policies are made available through a download page and users are required to affirmatively click a button to install an app Once in-stalled some apps further alert users to particular ways in which phone capabilities or data will be used with some permitting selective toggling (opting-in or opting-out) of certain features With DCTT apps in addi-tion to disclosures provided on a download page the user could be guided through a simple interactive module embedded in the app (such as is described above) in order to increase the chance of meaningful exposure to important information about the technology and how data will be handled At that point any user options such as those itemized above could be described and choices made
Opt-out Models
There are at least two different ways in which the term ldquoopt-outrdquo has been used in this context The conventional use of the term ldquoopt-outrdquo is characterized by an act which signals an individualrsquos intention to decline something that would have otherwise occurred without intervention A few others have used the term to refer to ldquorevocation of consentrdquo for example the United States COVID-19 Consumer Data Protection Act of 2020 Senate bill (S3663) would establish a default opt-in positionmdash
Ethics of Designing and Using DCTT 67
requiring ldquoaffirmative express consentrdquo for collection and use of prox-imity and other related datamdashand refers to individuals having a right to later revoke their consent through an ldquoopt-outrdquo The latter use of the term is not our focus here
Given this a DCTT app that is voluntarily downloaded through an affirmative act would be difficult to characterize as an opt-out approach This leaves more passive surveillance systems that rely on automatic in-stallation of self-activating technology onto users phones There are a range of views among the authors of this report about the value of an opt-out approach for DCTT with some arguing for an opt-out approach on grounds that it might increase coverage and would be ethically acceptable if accompanied by similar disclosures as above to ensure users are aware of the technology and data uses (Mello and Wang 2020) This approach would present users with a mechanism to opt-out if they wish which should be reasonably easy to effectuate Under these circumstances as noted above an ldquoopt-outrdquo would not be synonymous with mandating use of the technology
Others among the authors argue that there is reason to believe that opt-in approaches may be able to sufficiently achieve desirable levels of utilization relative to opt-out approaches Unfortunately data related to opt-in versus opt-out models of DCTT are very limited One recent sur-vey (Altmann et al 2020) found that across five countries (UK Germany France Italy US) slightly more people reportedly would download an app under an opt-in system (748) than would keep an app on their phone under an opt-out system (677) Moreover when US respondents were directly asked which approach they would prefer 60 indicated a preference for opt-in This remained true across various demographic variablesmdashgender region political affiliation lockdown status and other characteristics Whether actual behaviors would align with anticipated behaviors in the context of DCTT remains an unanswered question that should be carefully studied under real-world conditions There are a range of important empirical questions regarding how much and what kind of impact (positive or negative) various types of defaults might generate for public health and for different mobile phone user groups including vul-nerable and marginalized users
Opt-out models for app authorization may encounter greater legal
68 Digital Contact Tracing for Pandemic Response
and political challenges especially if the COVID-19 Consumer Data Pro-tection Act of 2020 (S3663) the competing Public Health Emergency Pri-vacy Act (S3749) or another similar bill is enacted in the United States Both of these standing bills require affirmative opt-in consent Opt-out approaches also risk negative reactions from some mobile phone users a small number of whom may go so far as to intentionally interfere with data because of the perceived intrusiveness of an automatically installed tracking platform (Dixit 2020)
Given these considerations and the apparent willingness of a large portion of the population to opt-in to use DCTT an opt-in approach to authorization should be instituted to accompany initial DCTT rollout The feasibility and value of opt-out approaches should continue to be evaluated informed by what is technologically possible what local as-sessments of benefits and harms of the technology reveal over time and our evolving understanding of the degree to which an opt-out approach is likely to increase or decrease utilization Opt-out approaches should not be precluded
Recommendations
bull A clear and concise module consisting of basic disclosure and vol-untary authorization should be developed to accompany DCTT This module should not take the form of ldquoclickwraprdquo terms of service or end-user agreements but rather provide only essential information necessary for an individual to make a decision More detailed disclosures (such as FAQs in plain language) should be made easily accessible to those who wish to learn more with no hidden surprises
bull An opt-in approach to authorization should be instituted to ac-company initial DCTT rollout The feasibility and value of opt-out approaches should continue to be evaluated informed by what is technologically possible what local assessments of benefits and harms of the technology reveal over time and our evolving under-standing of the degree to which an opt-out approach is likely to increase or decrease utilization among different populations Opt-out approaches should not be precluded
Ethics of Designing and Using DCTT 69
Promoting Equity and Fairness
Digital contact tracing technologies should be designed and used in ways that as far as possible promote an equitable distribution of benefits and burdens DCTT should be deployed in a manner that does not propa-gate preexisting patterns of unfair disadvantage or distribute harms and risks unfairly throughout the population For example communities with lower rates of technology and data access may benefit less from DCTT Special attention must be paid to communities that experience preexisting health disparities and to those that are being hardest hit by the pandemic
Digital Disparities
In the United States February 2019 data indicate that approximately 80 of the population are smartphone users (Pew Research Center 2020) though rates of mobile phone use are significantly lower among people over age 65 (53) people with any disability (58 2016 data) (Anderson and Perrin 2017) people with less than a high school edu-cation (66) people who earn less than $30000 per year (71) and people who live in rural areas (71) As a result these populations and communities may use DCTT in lower numbers thereby lessening the effectiveness of DCTT and the likelihood of benefit for these populations from such systems Moreover it has been reported that many older and less costly smartphones (roughly estimated at 10ndash20 of smartphones in the US) lack important capabilities required for the leading AppleGoogle platform to work (Bradshaw 2020) This is of special concern because some of the above groups that are less likely to own smartphones in general are also less likely to own newer smartphones with the needed capabilities Some within the above groups (eg people who are older and people identified as Hispanic African American or American Indian) are also disproportionately experiencing morbidity and mortality from COVID-19 (CDC 2020h)
One may argue that by using DCTT human and financial resources that would otherwise be spent on manual contact tracing will be pre-served and these resources can then be redirected to better meet the needs of those who are not otherwise being effectively served by the technology because of disparities or for other reasons This argument has intuitive
70 Digital Contact Tracing for Pandemic Response
appeal and should be taken seriously however it is unsettled whether DCTT will contribute sufficient efficiencies to the overall public health response to make it possible financially and logistically for manual ser-vices to be allocated in greater proportion to those who are unable to benefit from DCTT It is entirely possible that at least in the short-term DCTT may introduce new inefficiencies due to unintended consequences or the need for public health officials to follow up many more contacts One possible mitigation to the challenge of digital disparitymdashthough it does not solve the underlying challenge of ensuring net efficiency across systemsmdashmight be to provide mobile phones or other devices and data packages to those who would otherwise be left out
Disparate Risk of Harm from Surveillance and Data Gathering
Ensuring wide digital coverage does not however resolve other equity concerns It is important to consider that some populations may experi-ence greater harm and fear of harm from having their data collected For example some groups such as African Americans Hispanic Americans Muslim Americans and undocumented immigrants have more reasonable fear of their data being handed over to law or immigration enforcement and some groups have lower levels of trust in public health due to past injustices (CSM 2017 Pew Research Center 2017 Rodrigues et al 2018 Auxier et al 2019) Any data gathered by DCTT should be used solely for public health purposes Efforts should be made to assure members of these and other communities that their data will not be misused or made available to those outside of a public health context In addition if DCTT are used in the current pandemic this should be with the understanding that future use of DCTT in other contexts (eg law enforcement or im-migration enforcement) is presumptively unethical
Some preliminary polling related specifically to DCTT emphasizes the complexity of the challenges faced and the need for deeper public engagement (Anderson and Auxier 2020) The polling results suggest that people who identify as African American or Hispanic are more likely than people who identify as White to consider government tracking of mobile phones as acceptable These findings like many others are difficult to in-terpret given background political polarization on the issue More direct engagement is required to better understand how different communities comprehend and experience DCTT and other forms of surveillance
Ethics of Designing and Using DCTT 71
Discrimination and Stigma
Stigma may result from an individual being identified as COV+ or a neighborhood or establishment becoming identified as a ldquohotspotrdquo as a result of numerous COV+ people living in that area or having visited that establishment In particular certain groups may suffer more as a result of being associated with COVID-19 such as the well-documented blame that has been directed toward Chinese people (and broadly East Asian communities) or the communities that are disproportionately likely to contract the illness (Devakumar et al 2020) When identifiable lo-cation data are made public as has been the case in South Korea per-sonal and private information were revealed Furthermore businesses in South Korea that were identified as having patrons who tested positive for COVID-19 have suffered economic losses and stigma (N Kim 2020)
To avoid the stigma and potential discrimination that can result from being identified as COV+ DCTT must never make data publicly avail-able that could be used to identify persons who have tested positive Safe-guards must be in place to ensure that any identifiable data that may be gathered for public health purposes are protected If DCTT data are used to provide heat maps to the public of locations that COV+ individuals frequently visit so as to provide representations of geographic risk or for other reasons it is essential that care be taken to avoid unfairly distrib-uting further economic burdens or other stigmatizing and discriminatory outcomes
Recommendations
bull A commitment to equity means a commitment to ensuring that the benefits and burdens of DCTT are distributed fairly Public engagement is an important tool for assessing impact and to rectify inequities
bull States localities and institutions that recommend widespread use of DCTT should provide technology (eg mobile phones Blue-tooth devices) and free data packages to those who desire but lack access to these devices
bull If there are lower rates of adoption of DCTT systems in some identifiable communities public health authorities should iden-
72 Digital Contact Tracing for Pandemic Response
tify ways to compensate For example directing more non-DCTT resources and efforts toward those communities to meet specific needs that are elsewhere being supported by technology
bull If maps are generated based on DCTT to provide the public with the locations that COV+ individuals have visited steps must be taken to minimize the stigma and potential financial losses that could result from being identified as a hotspot
Instituting Transparent Governance and Oversight
DCTT must be developed with an eye toward both present and future implications Transparent and publicly trustworthy management gover-nance and oversight of DCTT technology and data is both a near- and long-term necessity We face significant uncertainties DCTT technologies are rapidly developing Their risks capabilities effectiveness and down-stream implications are not yet well understood
Concerns about ldquoSurveillance Creeprdquo
Significant concerns have been expressed by privacy advocates (Guari-glia 2020) and in the popular press (Giglio 2020) about what is known as ldquosurveillance creeprdquo Their worry is that state and corporate actors will use new surveillance technologies capacities and permissions well beyond the purposes for which they were initially justified to the public and beyond the time when they are useful for the COVID-19 pandemic
Surveillance creep should be guarded against Only those data that are necessary and relevant for the public health purposes at hand should be collected and used and data should be kept only for the period of time needed for those public health purposes Data should be used only for public health purposes
Any use of DCTT during the current pandemic would be justified by the circumstances of this pandemic and its use in this context does not set a precedent for future public health use (eg use in seasonal flu surveillance efforts) Future use will require independent justification Use of DCTT in the future in other contexts (eg law enforcement or immi-gration enforcement) is presumptively unethical
Broadly speaking efforts should be made to generate public aware-
Ethics of Designing and Using DCTT 73
ness and consensus that use of DCTT in COVID-19 efforts does not imply that future use is justifiable However generating this public aware-ness may be particularly challenging given the complexity of the informa-tional environment where public debate ranges from legitimate concerns about surveillance creep to conspiracy theories regarding the origins of the COVID-19 pandemic (Muller 2020) This means authorities bear spe-cial obligations to be clear on how they plan to use the technologies what oversight mechanisms will be employed to address potential abuse and how they intend to publicize the conditions under which programs will be terminated making sure they are followed
Oversight and Ethical Review
We are rapidly gaining knowledge about SARS-CoV-2 and COVID-19 but we still have essential gaps in our understanding In the United States public health responses including DCTT will generally be developed and coordinated by individual states regional consortia and associations (Reston Sgueglia and Mossburg 2020) Good governance in this con-text requires transparency and the creation of oversight bodies with the appropriate expertise and representation to allow nimble and effective responses while serving as trusted representatives
To address the range of ethics-related concerns about the design and use of DCTT digital surveillance oversight committees should be estab-lished perhaps at a state level and with a platform for national coordina-tion These committees can provide ethical and regulatory review prior to and concurrent with widespread use of DCTT These committees should be composed of a diverse group of experts capable of evaluating a DCTT system locally including members of communities that experience higher rates of digital disparity
When assessing the design and use of digital contact tracing systems these committees (and the public more widely) should consider not only the risks and benefits accrued during the COVID-19 pandemic but also implications for the future What kind of precedent might use of these technologies during the current pandemic set for future use capabilities in other infectious disease outbreaks or in other social contexts (eg law enforcement) How can we navigate safe use of these technologies in a way that preserves public trust in them and enables the possibility of future beneficial use
74 Digital Contact Tracing for Pandemic Response
As a start it should be emphasized that principles offered in this and other guidance documents do not apply only during the pandemic Future efforts to advance DCTT capabilities during quieter times should make every effort to follow them
Recommendations
bull Digital surveillance oversight committees should be established expeditiously with diverse and qualified membership to provide ethical and regulatory review prior to and concurrent with wide-spread use of a DCTT system
bull Understandable and publicly accessible rules must guide the col-lection access control use storage and combination of data by government authorities public and private institutions and other parties such as public health researchers
bull Only those data that are necessary and relevant for the public health response to COVID-19 should be collected and used
bull Identifiable data should be kept only for the period of time needed for the public health response to COVID-19
bull Identifiable data collected as part of this response should not be shared with anyone other than the relevant public health authorities without additional specific informed consent of individual users
bull Before a government or institution adopts a digital contact tracing program they should state the conditions under which the digital contact tracing program will be terminated
bull Future use of DCTT to advance public health or other efforts (eg use in seasonal flu surveillance) would require independent justification DCTT designed for public health use should not be used by law or immigration enforcement
bull The principles offered in this guidance document apply both during and following the COVID-19 pandemic
75
The implementation of digital contact tracing technology (DCTT) is likely to implicate a number of US laws at both the federal and state levels This section focuses primarily on federal laws as these laws apply nationwide and generally preempt conflicting state laws A comprehen-sive assessment of the legality of any particular DCTT program would require case-specific analysis and attention to relevant state laws includ-ing any that specifically address DCTT which may soon exist in one or more states The analysis here is limited to the United States foreign and international laws will not be addressed
Many of the laws discussed in this section are privacy laws designed to protect individuals from the harms that may result from the unautho-rized or improper use of their personally identifiable information (PII) Under these laws legal concerns will generally be minimized if privacy protections are built directly into the DCTT technology (eg ldquoprivacy by designrdquo) As a general principle DCTT should be designed to collect and store only as much PII as is necessary to achieve the public health purpose Collecting only proximity data for example is likely to raise fewer legal concerns than collecting both proximity data and geolocation data Likewise creating aggregated anonymized or de-identified data will raise fewer legal concerns than using and disclosing PII
As we have argued elsewhere in this guidance document however the public health and societal crisis caused by COVID-19 may justify
Legal Considerations
FOUR
76 Digital Contact Tracing for Pandemic Response
greater encroachments on individual privacy than would otherwise be permissible Regardless of the type of data collected privacy concerns will be reduced if users are afforded the right to choose whether their PII is collected and how it is used and disclosed As such DCTT should gener-ally secure meaningful user consent before collecting PII a process which typically requires both disclosure of relevant information and agreement on the part of the user
Privacy concerns will also be reduced if the use of PII is strictly lim-ited to tracking and limiting the spread of SARS-CoV-2 The use of DCTT data for other purposesmdashsuch as commercial or law enforcement pur-posesmdashwould raise additional legal and ethical concerns In addition DCTT developers may be required to implement governance policies that ensure the secure storage of PII limit data retention periods require transparency about data sharing and maintain records of responses to data requests from government authorities
In short the legality of a DCTT program under current United States law will depend on a number of factors including what type of data is collected how the data are used and who may access them how user consent is obtained whether the entity collecting and using the data is the government or a private corporation the context in which data are collected (eg employment education or commercial) and which states have jurisdiction over the program
Privacy law in the United States unlike in other jurisdictions such as the European Union (EU) and Australia is generally sector-specific and limited in scope The result is a patchwork of protections that differ significantly depending on the entity that collects the data and the type of data collected For example under current law telecommunication carriers are governed by different privacy rules than mobile broadband providers Given the complexity of existing federal privacy law we be-lieve that it would be beneficial for the US Congress to enact new privacy legislation that is specifically tailored to the use of DCTT in response to COVID-19 Congress appears poised at least to debate such legislation a pair of bills recently introduced in the Senate and one in the House of Representatives would significantly restrict the collection of PII by digital devices for COVID-tracing purposes S3663 S3749 HR 6866 116th Cong (2020)
Legal Considerations 77
Data Privacy and Data Security Laws
Telecommunications
A DCTT provider that collects data from a userrsquos mobile phone may be subject to the privacy rules governing telecommunication carriers which are enforced by the Federal Communications Commission (FCC) The data protected under these rules are limited however to certain types of PII termed ldquocustomer proprietary network informationrdquo (CPNI) More-over the rules generally apply only to telecommunications carriers and interconnected VoIP (Voice over Internet Protocol) providers
In particular under section 222 of the Communications Act of 1934 47 USC sect 222 and the implementing regulations of the Federal Com-munications Commission (FCC) telecommunications carriers and VoIP providers must establish and maintain systems designed to ensure that they adequately protect their subscribersrsquo CPNI and they are generally restricted from using or disclosing CPNI without the customerrsquos consent (unless the use of disclosure is needed to provide the services subscribed to by the customer) If customer consent is sought to use or disclose CPNI individual notice must be provided to the customer and such notice must provide sufficient information to enable the customer to make an in-formed decision as to whether to permit the requested use or disclosure
CPNI is individually identifiable information that carriers and pro-viders have collected about their customers including phone numbers called and the frequency duration and timing of such calls Of most relevance to DCTT a recent FCC Notice of Apparent Liability asserted that user geolocation data collected by mobile phone network carriers qualify as CPNI under sect 222 and related rules 35 FCC Rcd 1785 (2) (2020) Pursuant to this notice the FCC fined T-Mobile for selling to third parties location data that were derived from the communication between the mobile phones of T-Mobilersquos customers and nearby network signal towers (The FCC also levied fines against ATampT Verizon and Sprint on the same grounds (Valentino-DeVries 2020)) While the FCC has made its position clear that geolocation data are CPNI courts have yet to weigh in on the matter
Even if geolocation data are CPNI however the FCC can enforce sect 222 of the Communications Act only against telecom carriers and VoIP
78 Digital Contact Tracing for Pandemic Response
providers not against cable broadband and mobile broadband internet providers 47 USC sect 53(44) 47 CFR sect 93 In 2018 the FCC promul-gated a regulation stating that contrary to its prior position its sect 222 authority does not extend to cable broadband and mobile broadband internet providers Restoring Internet Freedom 83 Fed Reg 7852 (Feb 2 2020) (to be codified at 47 CFR pts 1 8 and 20) This regulatory shift was subsequently upheld by the DC Circuit Mozilla Corporation v Federal Communications Commission 940 F3d 1 (2019)
In addition to sect 222 the FCC has authority to regulate ldquocommon carriersrdquomdashincluding both telecommunication carriers and broadband internet providersmdashunder sect 201(b) of the Communications Act In the past the FCC has interpreted sect 201(b) to protect against ldquounjust and unreasonablerdquo privacy and data security practices with respect to custom-ersrsquo personal information beyond CPNI In 2016 the FCC promulgated a regulation asserting its authority under this interpretation However Congress overturned this regulation pursuant to the Congressional Re-view Act in 2017 SJ Res 34 115th Cong (2017) At present the extent of the FCCrsquos authority under sect 201(b) remains unsettled (Mulligan and Linebaugh 2019)
Consumer Protection
The collection storage release and transmission of digital user data in-cluding proximity contacts is more generally governed by the Federal Trade Commission (FTC) The FTC is an independent US law enforce-ment agency tasked with protecting consumers and promoting competi-tion across broad sectors of the economy (FTC 2020) The FTCrsquos primary legal authority with respect to consumer protection comes from Section 5 of the FTC Act which prohibits ldquounfair or deceptive acts or practices in or affecting commercerdquo 15 USC sect 45(a)(1) Note that the FTC and FCC have some overlapping authority to protect consumer privacy in the context of telecommunications (FCC and FTC 2017)
The FTC has interpreted Section 5 to require companies to be trans-parent and accurate about their collection of PII from consumers A com-pany may be found to have engaged in a deceptive practice if it fails to disclose that it is collecting user data or fails to disclose that it is sharing these data with third parties and to provide a general description of these third parties The FTC has used its authority under Section 5 numerous
Legal Considerations 79
times to discipline companies that purport in published privacy policies or other notices to provide protection for the privacy andor security of personal information yet fail to do so in practice For example the FTC may find it both ldquounfairrdquo and ldquodeceptiverdquo for a mobile app privacy policy to state that the app never discloses location information to third parties when in fact the app shares that information with the app developerrsquos service provider which in turn uses it to provide analytical data to the app developer that are used to create targeted advertising
The FTC does not use its Section 5 authority other than to protect consumers and generally does not consider ldquode-identifiedrdquo user data which are data that are not ldquoreasonably linkablerdquo to a consumer to be a subject for consumer protection In general data collected are not ldquorea-sonably linkablerdquo so long as the company collecting it ldquo(1) takes rea-sonable measures to ensure that the data are de-identified (2) publicly commits not to try to reidentify the data and (3) contractually prohibits downstream recipients from trying to reidentify the datardquo (FTC 2012)
Many states have laws that are similar to Section 5 prohibiting un-fair and deceptive acts and practices Both Section 5 and these similar state laws can be violated not only by misrepresentations (affirmative deception) but also by material omissions Thus a failure to inform an app user of the apprsquos collection of tracking data and the planned use and disclosure of those data could constitute a violation of these laws Com-panies providing DCTT apps should make sure that all such information is disclosed in the appsrsquo terms of use to which users must affirmatively agree
Childrenrsquos Online Privacy
Children who use DCTT may be protected by additional privacy protec-tions In particular collection of digital PII from children under the age of 13 is strictly regulated under the Childrenrsquos Online Privacy Protection Act (COPPA) (15 USC sectsect 6501ndash6505) Under COPPA PII includes ldquofirst and last name[] a persistent identifier that can be used to recognize a user over time and across different online services[] and geolocation infor-mation sufficient to identify street name and name of a city or town[]rdquo COPPA prohibits a website or online service from collecting personal information (including location information) from children under age 13 without obtaining verifiable parental consent Note that there may be an
80 Digital Contact Tracing for Pandemic Response
exception to this requirement for an ldquoinvestigation on a matter related to public safetyrdquo 16 CFR sect 3125(c)(6)(iv)
Electronic Surveillance
In addition to misuse of user data by DCTT providers another privacy concern is that a third party may be able to access sensitive PII that is collected and stored by a DCTT system without the userrsquos knowledge and consent There are a number of federal criminal laws however that would likely prohibit such unauthorized access to PII
In particular the Electronic Communications Privacy Act of 1986 (ECPA)mdashwhich includes the Wiretap Act (18 USC sectsect 2510ndash2522) the Stored Communications Act (18 USC sectsect 2701ndash2711) and the Pen Register Act (18 USC sectsect 3121ndash3127)mdashmakes it a crime to access elec-tronic communications without authorization Individuals who violate the ECPA face up to five years in prison and fines up to $250000 Victims are also entitled to bring civil suits and recover actual damages in addi-tion to punitive damages and attorneyrsquos fees for violations
Generally the access restrictions in the ECPA apply unless consent is given or if access is authorized by statute for law enforcement purposes For example an employer is generally forbidden from accessing an em-ployeersquos private emails However if consent is given in the form of an employment contract that explicitly authorizes the employer to access emails it may be lawful under the ECPA for the employer to access such information Along the same lines the ECPA would likely prohibit an employer from accessing contact tracing data on an employeersquos phone without the employeersquos consent However the ECPA would likely not prohibit duly authorized government public health officials from access-ing contact tracing data without consent
As its name suggests the Stored Communications Act (SCA) regu-lates access to communications at rest that is not in transit The SCA makes it unlawful to intentionally access a facility in which electronic communication services are provided and to obtain alter or prevent au-thorized access to a wire or electronic communication while it is in elec-tronic storage in such a system As such the SCA would likely apply only to centralized collection of contact tracing data
The Pen Register Act covers any ldquosignaling informationrdquo exchanged in a communication such as phone numbers The statute does not reach
Legal Considerations 81
the content of such communications however An expansive interpreta-tion of the Pen Register Act would cover Bluetooth ldquohandshakesrdquo as they are merely signaling information between devices which do not carry content See United States v Forrester 512 F3d 500 (9th Cir 2007) (find-ing that email headers and IP addresses are akin to pen registers and have no Fourth Amendment protection) Unlike the SCA there is no statutory exclusionary rule that applies when the government illegally uses a pen register trap and trace device Additionally there is no private cause of action against the government for violations of the Pen Register Act
State Data Privacy Laws
States have a variety of privacy laws and are increasingly seeking to reg-ulate the online collection of personal information and the use and dis-closure of such information To date most of these laws focus more on transparency and protection from unauthorized access than on restricted collection and use (except with respect to biometric information) seek-ing to ensure that individuals who use websites or online services such as mobile applications do so on an informed basis with respect to the privacy provided by those sites and services Two examples of such state laws are the California Online Privacy Protection Act (CalOPPA) and the California Consumer Privacy Act (CCPA) Both laws require notice to in-dividuals who use websites or online services such as mobile applications in order to ensure that users are informed about the privacy of personal information collected by those sites and services (The CCPA also applies to data collection off-line) Both laws treat IP addresses and location data as types of potentially identifiable personal data and so would very likely apply to DCTT apps used by California residents
CalOPPA requires that the operator of any website mobile appli-cation or other online service (ldquoSiterdquo) post a privacy policy on the Site disclosing certain information regarding the Sitersquos collection use and dis-closure of PII CalOPPA applies to any Site that is accessible to California residents The required disclosures are not onerous and would apply only to collection of data that are identifiable to an individual person (but depending on who collects the data location data together with a device identifier are identifiable to the user)
The CCPA requires that any entity qualifying as a ldquobusinessrdquo provide its ldquoconsumersrdquomdashdefined as lawful residents of Californiamdashwith specific
82 Digital Contact Tracing for Pandemic Response
disclosures about the businessrsquos collection use and disclosure of personal information Importantly the CCPA applies only to for-profit businesses that meet certain thresholds of revenue or access to consumer informa-tion A public health agency or a nonprofit organization would not be subject to the CCPA Cal Civ Code sect 1798140(c)
The CCPA defines ldquopersonal informationrdquo as ldquoinformation that iden-tifies relates to describes is reasonably capable of being associated with or could reasonably be linked directly or indirectly with a particular consumer or householdrdquo The statute provides a nonexclusive list of po-tential identifiable personal information including ldquogeolocation datardquo In accordance with the CCPA businesses must provide consumers with a notice ldquoat or before the point of collectionrdquo of personal information which must describe the personal information to be collected and the pur-poses for collecting that information Businesses must additionally allow consumers to request access to and request deletion of personal informa-tion Businesses must allow for consumers to opt-out of the sale of any personal information Developers of COVID-tracing apps would want to build in compliance with these requirements In addition California Civil Code sect 1798815(a)(1) requires companies to ldquomaintain reason-able security procedures and practices appropriate to the nature of the information it processesrdquo
Like privacy laws generally the CCPA does not grant consumers rights regarding the use of de-identified information However the CCPA does require businesses to implement processes that prohibit re-identification of de-identified information as well as technical safeguards to prevent inadvertent release of that information Cal Civ Code sect 1798140(h)
Health Information Privacy
Many DCTT systems will be designed to collect health-related data of users such as symptom tracking SARS-CoV-2 test results and prior ex-posure to a person who is COV+ Individuals may have additional privacy protections with respect to the use and disclosure of this health-related information
The use and disclosure of individually identifiable health information is strictly regulated under the privacy and security rules implementing the
Legal Considerations 83
Health Insurance Portability and Accountability Act (HIPAA) HIPAA is limited in application however to health care providers and health insur-ance plans (ldquocovered entitiesrdquo) and ldquobusiness associatesrdquo of such entities ldquoBusiness associatesrdquo under HIPAA are persons who perform services for covered entities and need access to personal health information to do so
HIPAA-covered entities must have written authorization to use or disclose identifiable health information (ldquoprotected health informationrdquo or PHI) from the individual to whom such information pertains unless the HIPAA regulations promulgated by the US Department of Health and Human Services (HHS) provide an exception to the requirement for such individual authorization
Among the exceptions to the individual authorization requirement is an exception for certain uses and disclosures of PHI for public health purposes 45 CFR sect 164512(b) This exception would permit for exam-ple a HIPAA-covered entity to disclose the PHI of an individual who tests positive for SARS-CoV-2 to a public health authority A ldquopublic health authorityrdquo is an agency or authority of the US government a state ter-ritory a political subdivision of a state or territory or Indian tribe that is responsible for public health matters as part of its official mandate as well as a person or entity acting under a grant of authority from or under a contract with a public health agency such as a contact tracer Id sect 164501
Many DCTT developers are HIPAA business associates and any use and disclosure of PHI collected through DCTT used on behalf of HI-PAA-covered entities is restricted under the HIPAA privacy rules Nota-bly in response to COVID-19 HHS announced that its Office for Civil Rights would exercise its enforcement discretion and would not impose penalties for violations of certain provisions of the HIPAA Privacy Rule against health care providers or their business associates for the good faith uses and disclosures of protected health information for public health and health oversight activities during the nationwide public health emergency 85 FR 19392 (2020)
Many states also have health information privacy laws The HIPAA privacy rule sets a ldquofloorrdquo of privacy protections allowing the states to be more protective of privacy More specifically HIPAA preempts a state law if (but only if) the state law is ldquocontraryrdquo and less protective of privacy than the HIPAA privacy rule However if a state law is determined by the
84 Digital Contact Tracing for Pandemic Response
Secretary of HSS to be necessary to serve a ldquocompelling need related to public health safety or welfarerdquo it may survive preemption even if it is less privacy-protective than HIPAA 45 CFR sect 160203 (a)(1)(iv)
The Public Health Service Act also restricts the use of certain per-sonally identifiable information collected by entities involved with public health activities without the individualrsquos consent 42 USC 242m(d)
Labor and Employment Privacy Rights
Labor and employment lawsmdashthat is laws that govern the relationships between employers and employeesmdashmay prove relevant to DCTT espe-cially if employers mandate the use of DCTT or seek to collect health information regarding their employees using DCTT Depending on the built-in privacy protections of the DCTT system an employer may be able to access important health information from an employeersquos phone As noted above the ECPA would generally prohibit an employer from ac-cessing this information without the employeersquos consent Even with con-sent however there are limits on the collection and use of an employeersquos health information
In particular the use of DCTT may raise special concerns about em-ployment discrimination for example if an employer were to fire an em-ployee who tests positive for SARS-CoV-2 (COV+) or who has a known SARS-CoV-2 exposure The Americans with Disabilities Act (ADA) pro-tects disabled employees from discrimination and restricts the collection of personal health information by employers The Equal Employment Opportunity Commission (EEOC) which is the federal agency tasked with enforcing the ADA in the employment context would likely con-sider COV+ to be a ldquodisabilityrdquo under the ADA and analogous state laws prohibiting discrimination against disabled people COV+ is likely to be a ldquodisabilityrdquo especially where the individual is symptomatic andor experi-ences related health issues or if it is later determined that testing positive for SARS-CoV-2 leads to long-term or chronic health effects ldquoExposure to a COV+ personrdquo could also be covered by those laws because a person exposed to a COV+ individual could well be perceived as being disabled by being considered likely to be infected
Legal Considerations 85
The ADA generally requires that businesses make ldquoreasonable accom-modationsrdquo for persons who are disabled which may include individuals who are COV+ or who have a preexisting disability that places them at higher risk from or may be exacerbated by COVID-19 The EEOC has published guidance on reasonable accommodations under the ADA and related laws in the context of COVID-19 (EEOC 2020) Among other things this guidance clarifies that consistent with the ADA employers may take temperatures or otherwise collect health information about employees during the pandemic crisis so long as they keep that infor-mation confidential As of May 18 2020 the EEOC has not provided guidance that specifically addresses the applicability of the ADA to the use of DCTT by employers
In addition employment laws such as the ADA and the Family and Medical Leave Act (FMLA) and state law equivalents generally limit disclosure of information and require employers to keep confidential any employee personal health information related to a disability or request for medical leave Under the ADA any information regarding the medical condition or history of an employee that an employer obtains as part of an examination or inquiry into a disability could constitute a confidential medical record that can be disclosed only to certain individuals in lim-ited circumstances 42 USC sectsect 12112(d)(3)(B) and 12112(d)(4) The FMLA also prevents the disclosure of records related to medical histories in connection with an employeersquos leave request or eligibility 29 CFR sect 825500(g) The EEOC and some courts have gone further and taken the position that any information concerning an employeersquos medical con-dition is protected under the ADA or FMLA
As discussed elsewhere in this guidance document employers may have a good reason to employ DCTT in order to ensure workplace safety and limit the spread of SARS-CoV-2 in the community Employers may also face legal liability if they fail to protect employees (or customers) from potential exposure or infection In particular employers have an obligation under the Occupational Safety and Health Act to keep the workplace safe for employees In response to COVID-19 the Occupa-tional Safety and Health Administration (OSHA) has developed guid-ance on preparing the workplace (OSHA 2020) The CDC has also pre-pared guidance on healthy business operations and reducing the spread
86 Digital Contact Tracing for Pandemic Response
of SARS-CoV-2 in the workplace (CDC 2020c) Employers must strike an appropriate balance between avoiding employment discrimination and promoting workplace safety
Reflecting the need for such a balance the employee protections un-der the ADA and other employment laws are not absolute and are limited by among other things the need to protect the health and safety of other employees and the public Protection for workplace safety and health generally will justify appropriately tailored measures such as inquiries into an employeersquos personal health status or whether someone has tested positive for SARS-CoV-2 temperature checks and removal of employees from the workplace who are experiencing symptoms or have tested posi-tive and have not been cleared to return to work
Note finally that the use of DCTT by employers should be evaluated in conjunction with the hazard pay sick leave and other benefits that are available to employees Under the Families First Coronavirus Response Act employers with more than 50 employees and fewer than 500 employ-ees are required to provide two weeks of paid sick leave to an employee who stays home because of COVID-19 Pub L No 116-127 134 Stat 178 (2020) This paid leave extends to those who are themselves ill are quarantined or are awaiting a diagnosis as well as those who are caring for sick family members However reporting suggests that more than 75 of US workers will not qualify for benefits under this act (Cochrane Miller and Tankersley 2020)
Constitutional Privacy Rights
A DCTT program involving only private actors operating on the ba-sis of voluntarily provided information would not present constitutional privacy issues But any government-directed use of digital technology to support public health tracking and contact tracing involving mandatory government surveillance may potentially implicate a variety of consti-tutional protections These constitutional protections apply to actions taken by any level of government in the United States While state gov-ernments have broad policing powers in the area of public health (Jacob-son v Massachusetts 197 US 11 (1905)) and are generally allowed to enforce legislation not preempted by federal laws even emergency and
Legal Considerations 87
health-protective laws must be consistent with the US Constitution (HHS 2019 CDC 2020f)
Fourth Amendment Search and Seizure
Many people considering whether to use a DCTT app may be concerned that government enforcement agencies would obtain tracing data and use those data to conduct criminal prosecutions or immigration proceed-ings Constitutional protections notably the Fourth Amendmentrsquos limit on warrantless searches limit the governmentrsquos use of personal data in the criminal context However exceptions exist allowing law enforce-ment to access information even when such access would generally be prohibited How the government accesses personal data stemming from contact tracing needs to be scrutinized and protections will hinge on the manner of access
In general the Fourth Amendment protects ldquo[t]he right of the peo-ple to be secure in their persons houses papers and effects against un-reasonable searches and seizuresrdquo As originally interpreted the Fourth Amendment was considered tied to common-law trespass That is no lon-ger the case US Supreme Court precedent interprets the Fourth Amend-ment to protect ldquopeople not placesrdquo and extends to the protection of certain expectations of privacy such as location information as long as such expectations are reasonable Katz v United States 389 US 347 351 (1967) A warrantless government search is unconstitutional when the information sought is private and such expectation of privacy is ldquoone that society is prepared to recognize as reasonablerdquo Smith v Maryland 442 US 735 743ndash44 (1979)
The constitutionality of a search will revolve around the following analysis whether the digital program either violates an individualrsquos ldquorea-sonable expectation of privacyrdquo (likely triggered by programs collecting large amounts of location andor health data) or involves a government ldquotrespassrdquo (likely triggered by required app downloads) Katz v United States 389 US 347 (1967) United States v Jones 565 US 400 (2012)
Courts will most likely weigh the intrusiveness of the measures taken in implementing a search standard against the severity of the situation governmental and individual interests and accountability measures and safeguards built into the system
Voluntary sharing by individuals of their information with other par-
88 Digital Contact Tracing for Pandemic Response
ties including the government would mean that there was no reasonable expectation of privacy and would not raise the issues elaborated above It is worth noting that consent may not be considered voluntary if coerced or conditioned especially with regard to public employees or students of public institutions
Third-Party Doctrine
Some legal doctrines allow for the governmentrsquos acquisition of otherwise private information consistent with Fourth Amendment privacy protec-tions The third-party doctrine for example provides that individuals have no reasonable expectation of privacy in information voluntarily shared with others even if the information is revealed on the assumption that it will be used only for a limited purpose and the confidence placed in the third party will not be betrayed Smith v Maryland 442 US 735 (1979) United States v Miller 425 US 435 (1976) This applies to in-formation provided by third parties (mobile carriers internet service pro-viders medical tracking device manufacturers etc) to the government under order or request even when the third partyrsquos end-user agreements or privacy policies create an expectation of privacy
The Supreme Court has narrowed the applicability of the third-party doctrine to exclude use and disclosure of ldquohistoricalrdquo cell-site location information (CSLI) data For example in Carpenter v United States 138 S Ct 2206 (2018) the Court reasoned that the third-party doctrine does not justify use and disclosure of historical CSLI because an individual does not provide that information voluntarily Rather that information is pervasively collected by the cell phone company without any affirmative action on the part of the individual The Court did not express a view on ldquoreal-timerdquo CSLImdashlocation information that live-tracks a cell phonersquos locationmdashor on GPS data that may be stored in the phone itself
The Carpenter decision builds on a line of cases related to searches of digitally stored location data In Riley v California 134 S Ct 2473 (2014) the Court held that absent exigent circumstances law enforce-ment must obtain a warrant to search an individualrsquos phone Exigent circumstances are those that require immediate action because there is a probability that evidence may be destroyed The use of a centralized data-base for collection of digital contact tracing data would obviate deletion
Legal Considerations 89
concerns If the data are stored locally in the phone issues may arise as to whether law enforcement may suspect the data may be deleted following an arrest
Similarly in United States v Jones 132 S Ct 945 (2012) Justice So-nia Sotomayor authored a concurring opinion arguing that the use of a GPS to track a defendantrsquos whereabouts has the potential of providing the government with enough data points to create a ldquomosaicrdquo of the personrsquos life Location data obtained through centralized location contact tracing have the potential of providing information on an individualrsquos where-abouts beyond whatrsquos necessary for determining proximity to infected individuals Localized data may also raise the same issues if accessed by law enforcement
Following Carpenter several courts have addressed the constitution-ality of novel location tracking In Massachusetts for instance a federal district court concluded that police use of a ldquopole camerardquo on a utility pole to investigate the movements of an individual constituted a search under the Fourth Amendment United States v Moore-Bush 381 FSupp3d 139 (D Mass 2019) The court reasoned that even in a public space an in-dividual still retains a reasonable expectation of privacy ldquoin the whole of their physical movementsrdquo Citing Carpenter and Jones the court stated that the governmentrsquos unrestrained power to collect data that reveal pri-vate aspects of identity is susceptible to abuse and gives police access to a category of information that is ldquootherwise unknowablerdquo Long-term monitoring of a personrsquos movements consequently violates that individ-ualrsquos expectation of privacy Notably the court emphasized the capability of the camera to create a searchable digital log of the photos taken for the eight-month period during which the camera was used
State courts have also weighed in on the issue The Massachusetts Supreme Judicial Court found that police access to real-time location data pinpointing an individualrsquos movement whether from a third party or a cell-site simulator infringes upon an individualrsquos reasonable expec-tation of privacy Commonwealth v Almonor 120 NE3d 1183 1195 (Mass 2019) The Washington Supreme Court for its part held that a cell phone ping used to locate the defendantrsquos vehicle in real time is a search under the Fourth Amendment requiring a warrant absent exigent circumstances State v Muhammad 428 P3d 1177 (2018) And the Colo-
90 Digital Contact Tracing for Pandemic Response
rado Court of Appeals held that police use of a video pole camera to con-tinuously surveil a defendantrsquos fenced-in backyard constitutes a search under the Fourth Amendment People v Tafoya 2019 BL 457321 Colo Ct App 17CA1243 (2019)
Application of Carpenter by lower courts to novel location-tracking tactics is still evolving and it is as yet unclear how the narrower interpre-tation of the third-party doctrine will continue to be expanded and ap-plied particularly in cases of short-term monitoring of massive amounts of location andor health data Moreover it is unclear whether Carpenter would apply to DCTT data collected by the government itself
Special Needs Doctrine
An argument in favor of the constitutionality of government DCTT programs is that the ldquospecial needsrdquo doctrine would apply Under this doctrine a warrantless search that would otherwise violate the Fourth Amendment might be permissible based on a special need relating to pub-lic health When the search is conducted for a nontraditional law enforce-ment purpose and circumstances make securing a warrant impracticable the Supreme Court has ruled that warrantless searches may be permissi-ble The special needs doctrine however is highly controversial because it is not a consistently applied Fourth Amendment exception so it is diffi-cult to predict when courts would authorize nontraditional surveillance Some factors considered by the court are (1) the balance between the intrusiveness of the government action and the anticipated public bene-fits (2) the existence of legislative authorization (3) judicial process or the ability of the subject individual to challenge the government action (4) the scope or breadth of government action and (5) the likelihood of the collected data being used in criminal proceedings The Supreme Court did note in Chandler v Miller 520 US 305 (1997) that a ldquorisk to public safety [that] is substantial and realrdquo may justify ldquoblanket suspicionless searches calibrated to the riskrdquo citing as examples the routine searches conducted at airports and entrances to some official buildings (Searches within the context of immigration are further analyzed below)
Immigration Enforcement
Exceptions apply to the constitutional requirement that a warrant ac-company an unreasonable search or seizure in the immigration context
Legal Considerations 91
For example an exception to the general warrant requirement is the bor-der search exception which allows government officials to search and seize without a warrant persons and property at the border or at the functional equivalent of a border See United States v Montoya de Her-nandez 473 US 531 (1985) United States v Flores-Montano 541 US 149 (2004) Federal regulation authorizes immigration officials to oper-ate within 100 miles of any US external boundary (See 8 CFR sect 2871 defining ldquoreasonable distancerdquo as ldquowithin 100 air miles from any external boundary of the United Statesrdquo) A functional equivalent of a border may include any airport where international flights may be received automo-bile checkpoints servicing international traffic and vessels in territorial waters Government officials however must still have ldquoreasonable suspi-cionrdquo of an immigration violation or a crime to search or seize persons or property
In the context of digital data Customs and Border Protection (CBP) officials may conduct either manual or forensic searches of electronic devices at the border or its functional equivalent A manual search is considered a routine search and may include accessing the phone and ldquobrowsingrdquo its contents If the electronic device is password protected individuals must provide information for unlocking the device Forensic searches on the other hand are nonroutine and involve a more invasive search of the electronic devicersquos contents Federal circuit courts are split on whether a CBP agent needs ldquoreasonable suspicionrdquo before conducting a forensic search of an electronic device But Supreme Court precedent clearly states that suspicionless searches are not unconstitutional when public safety is considered Skinner v Ry Labor Execsrsquo Assrsquon 489 US 602 (1989)
A recent CBP directive provides guidance and standard operating procedures regarding forensic searches of electronic devices CBP 3340-049A Border Search of Elec Devices (DHS 2018) The directive states that CBP officers may detain electronic devices or copies of the informa-tion contained within these devices for a reasonable period time not to exceed five days This directive raises the concern that travelers may be required to turn over contact tracing data stored on their phone to CBP officers Note that the directive has been challenged in federal court and is currently awaiting appeal Alasaad v Nielsen 419 FSupp3d 142 (D Mass 2019)
92 Digital Contact Tracing for Pandemic Response
Searches in Schools
Another exception to the general warrant requirement applies to searches by non-law-enforcement government officials in public schools (ie school officials) Within this context school officials have broad powers to conduct searches as long as those searches are reasonable Searches by individuals in private schools are not governed by the Fourth Amend-ment State regulation of searches in private schools varies (See US DOE 2009)
Related Federal Privacy Statutes
Outside the Fourth Amendment context certain laws provide protections against government collection of and access to personal data The USA Freedom Act of 2015 for example bans the governmentrsquos bulk collec-tion of internet metadata and telephonic records which was previously allowed under Section 215 of the USA Patriot Act The government must now identify with specificity the identity of a person account address or personal device when requesting records The law allows for the acqui-sition of data by two degrees of separationmdashor ldquohopsrdquomdashfrom targeted individuals If a centralized system in contact tracing is used it is unclear whether the government may need to resort to this provision since it would likely have consent from individuals to collect and use the data
The Privacy Act of 1974 also regulates the collection use and disclo-sure of personal data but applies only to federal agencies (and their con-tractors) not to state or local agencies 5 USC sect 552a The Act protects against disclosure of individually identifying ldquorecord[s]rdquo that are kept within a ldquosystem of recordsrdquo The Act limits disclosure of information ldquoexcept pursuant to a written request by or with prior written consent of the individual to whom the record pertainsrdquo Certain disclosures are ex-empt from the Actrsquos applicability Pertinent disclosure exceptions are for records required to be disclosed under the Freedom of Information Act (FOIA) or disclosures ldquoto a person pursuant to a showing of compelling circumstances affecting the health or safety of an individualrdquo A disclosure under FOIA however would not include information in ldquopersonnel and medical files and similar filesrdquo when disclosure ldquowould constitute a clearly unwarranted invasion of personal privacyrdquo FOIA Guide 2004 Edition
Legal Considerations 93
Exemption 6 If non-anonymized data are turned over to the federal or state governments it is important to consider whether PII would be pro-tected from disclosure under FOIA or state freedom of information laws
Consent
User consent is a cross-cutting issue for evaluating many of the laws and regulations governing personal information privacy discussed in the prior sections In general privacy laws can be justified on the grounds that an individual should have the option to control with various types and de-grees of limitation the collection use retention andor disclosure of in-formation pertaining to that individual by others As such many privacy laws start from the premise that absent an individualrsquos consent use or disclosure of that individualrsquos PII is impermissible except for certain enu-merated purposes deemed to outweigh the individualrsquos privacy interests
Consent like agreements in general can be manifest in different ways in specific circumstances In some cases an affirmative actionmdashsuch as a signaturemdashis needed to demonstrate consent In other cases inactionmdashsuch as declining to ldquounsubscriberdquo from receiving certain unsolicited emailsmdashconstitutes consent Where a law requires a ldquowrittenrdquo signature in all but a few contexts the signature may be executed electronically In the United States that means the ldquosignaturerdquo may consist of any of the following ldquoan electronic sound symbol or processrdquo so long as it is ldquoattached to or logically associated with a contract or other record and executed or adopted by a person with the intent to sign the recordrdquo Electronic Signatures in Global and National Commerce Act (E-SIGN) (15 USC 7006)
The scope of a consent depends on what was deemed to be under-stood by the consenting party That is least clear when the consent is in-ferred from inactivity even if terms stating the consequences of inactivity have been provided The scope of consent is most clear when the terms agreed to have been presented to or provided by the consenting party in a conspicuous documented manner and a record exists of those terms Courts uphold the validity of clickthrough agreements because users are deemed to review the terms to which they respond by clicking ldquoI agreerdquo
94 Digital Contact Tracing for Pandemic Response
But where terms are ambiguous or confusing buried in other text or presented obscurely the ldquoI agreerdquo action may not mean the user actually agreed to specific terms
Terms of Use and Privacy Policies for apps are often written in com-plicated or nuanced language with key points difficult to discern More-over they are generally hard to read on a mobile device Many users of mobile phone apps agree to such terms without even attempting to read or to understand them As such it is often questionable whether an app user has knowingly agreed to all the terms of those documents Presen-tation of terms in large typeface short sentences simple language and direct disclosures makes user consent more meaningful
For contact tracing apps that collect PII andor PHI consent will overcome the restrictions of many if not most privacy laws provided the consent is freely given reflects a full understanding of the terms for use collection and disclosure of the information and is confirmed by an af-firmative act such as a click that may be executed only upon a complete reading of Terms of Use and Privacy Policies Whether consent may be deemed ldquofreely givenrdquo in certain circumstances depends on contextual understandings of party relationships including the employer-employee and government-citizen relationships
Anti-discrimination and Individual Freedom Laws
Any measure taken to protect public health and safety must comply with the Constitution and civil rights laws such as the ADA that prohibit discrimination against persons in certain protected categories such as race gender religion or disability In addition certain implementations of DCTT could be challenged under a variety of individual freedom protections
Anti-discrimination Laws
In general it would be impermissible to use DCTT in a way that either targets or excludes people on the basis of their membership in one of these protected categories
When motivated by animus against a protected class as defined by law and not narrowly tailored to advance a compelling government inter-
Legal Considerations 95
est a discriminatory regulation would be considered unconstitutional un-der the Equal Protection Clause of the Fourteenth Amendment to the US Constitution See Jew Ho v Williamson 103 F10 CCND Cal (1900) (striking down a quarantine imposed by San Francisco in response to an outbreak of bubonic plague because it was racially motivated) see also Church of Lukumi Babalu Aye v Hialeah 508 US 520 (1993) (supposed public health measure unconstitutional because it targeted the practices of one religion)
The risk of unintentional yet illegal discrimination in using DCTT is a real possibility Recent studies of infection rates among the population have revealed a larger-than-proportional infection rate among certain minority communities such as Latinx African American and American Indian communities (NYC DOH 2020) Programs that target specific ra-cial ethnic tribal or religious groups may raise constitutional and other legal concerns
Religious Freedom Laws
The use of DCTT may also raise concerns about religious freedom For example there may be religious objections to restrictions on gathering for worship carrying a mobile phone or the use of imaging technology Under current Supreme Court precedent generally applicable laws that do not discriminate against religion on their face do not violate the Free Exercise Clause of the First Amendment even if those laws have an inci-dental effect on the exercise of religion Employment Div Dept of Hu-man Resources of Oregon v Smith 494 US 872 (1990) These laws need not be justified by compelling government interest (the ldquostrict scrutinyrdquo standard of review) the government need only show that they are ratio-nally related to a legitimate interest On the other hand laws that are not neutral and not of general applicability must be justified by compelling government interest and must be narrowly tailored to advance that in-terest if it burdens religious practicesmdasha very tough hurdle to overcome Church of Lukumi Babalu Aye v Hialeah 508 US 520 (1993)
This general approach however is disrupted in some contexts by statutes adopted to provide greater protection to religious freedom The federal Religious Freedom Restoration Act (RFRA) requires strict scru-tiny for federal actions that burden religion and many states have ad-opted ldquostate RFRAsrdquo that do the same for actions by state and local
96 Digital Contact Tracing for Pandemic Response
governments The Religious Land Use and Institutionalized Persons Act which extends similar protections to persons confined to an institution such as a prison jail or mental health facility may also be relevant 42 USC sect 2000cc
Under either standard of review courts will examine whether a gov-ernment action imposes a substantial burden on religious exercise if not no religious freedom violation has occurred Such a finding is unlikely for DCTT programs absent evidence that the government is using the digital information to take action against religious persons that is not necessary to avoid the spread of a serious disease Nevertheless legal challenges on religious freedom grounds cannot be ruled out
Legislative Recommendations
bull Congress should enact new legislation specifically tailored to fa-cilitate the use of DCTT as part of the public health response to COVID-19 while also protecting user privacy and ensuring data security
bull Congress should require DCTT developers to disclose to users in clear language the nature of the information that would be col-lected how it would be collected how it would be stored and for what purposes it may be used
bull While the rollout of DCTT should initially employ an opt-in au-thorization approach the feasibility acceptability and value of opt-out approaches should continue to be evaluated As such opt-out approaches to consent should not be precluded by legislation
bull Congress should prohibit the commercial use of data collected for COVID-19 response by DCTT
bull Congress should prohibit discrimination on the basis of data col-lected by DCTT
bull If Congress is unable to enact suitable legislation state legislatures should work toward enacting similar laws for their jurisdictions A ldquomodelrdquo state law should be rapidly developed to facilitate na-tionwide uniformity of legal requirements
97
Public Health
Characteristics That Make Data Useful to
Public Health for Reducing Disease Transmission
bull Technologies or apps with the goal of enhancing public health ca-pacity to identify cases and trace contacts in order to control the spread of SARS-CoV-2 should be designed to match functionality with that goal
bull Technologies or apps may produce some false negatives or false positives but they should be accurate enough that public health authorities feel confident that they support and donrsquot detract from contact tracing efforts
bull DCTT approaches for public health should be designed to facili-tate the following
deg Identifying contacts including those who may not be easily found otherwise
deg Finding and notifying contacts rapidly before they develop symptoms if infected
deg Analyzing the nature of contact to determine whether con-tact is high medium or low risk to support decisions about whether quarantine should be mandatory should be volun-tary or is not needed
deg Following up with cases and contacts so that public health can provide resources to support isolation and quarantine at home
Recommendations
FIVE
98 Digital Contact Tracing for Pandemic Response
bull Data collected through DCTT should be made available to public health professionals and to researchers in de-identified form to support population-level epidemiologic analysis
Ethics
Collecting Data to Inform Policy and Practice
bull Reviews of DCTT systems must be conducted in part by an inde-pendent intermediary that has established the publicrsquos trust
bull Those who authorize use of DCTT within a particular jurisdic-tion or institution should continuously and systematically monitor the technologyrsquos performance in that context This should include monitoring for effectiveness and benefit monitoring for harms and monitoring for the fair distribution of both benefits and harms They should also monitor evidence that is being generated in other contexts about their selected technological solution and about other competing technologies
bull Data should be available to users that would permit them to further investigate their personal risk with public health officials or other health workers to add a layer of protection against unnecessary quarantine
Public Trust and Public Attitudes
bull More research into public attitudes is needed In particular in-depth qualitative research should examine public attitudes about perceptions of trust in DCTT among different communities which features of DCTT influence trust and the extent to which people are willing to provide different types of data through DCTT to help their community
bull States and localities that are considering adopting DCTT should engage with the public to increase their understanding of the acceptability of DCTT design features and uses among diverse communities
Recommendations 99
Designing Flexible Technology to Maximize
Public Health Utility While Respecting Other Values
bull Technology companies should not alone control the terms condi-tions or capabilities of DCTT nor should they presume to know what may be acceptable to members of the public
bull A ldquovalues in designrdquo approach to development of DCTT should be adopted (Flanagan Howe and Nissenbaum 2008 Knobel and Bowker 2011) Robust public- and user-engagement activities should be pursued to identify and incorporate to the extent pos-sible a range of values into the design of the technology These values may include privacy but also autonomy efficiency equity or others Technology design should reflect an appropriate balance and prioritization of identified values
bull Technology design should not be static but should be capable of evolving depending upon local conditions new evidence and changing preferences and priorities
bull DCTT should be designed to have a base set of features that pro-tect privacy with layers of additional capabilities that users may choose to activate An initial default should be that user location data are not shared but users should be provided with easy mech-anisms and prompts to allow for opting-in to this capability with encouragement to the public if and as it is shown to be critical to achieving public health goals
Policy Positions to Advance Widespread
Use of Digital Contact Tracing Technologies
bull DCTT use should not be mandated at this time given uncertainty about potential harms and benefits Additional technology user and real-world testing is needed
bull Incentives can be a useful complement to encouragements how-ever any incentives for users to install and use DCTT must be equitable should not be coercive and should align with effective use of the technology
100 Digital Contact Tracing for Pandemic Response
bull Trusted leaders should be enlisted to communicate effectively with the public about DCTT and encourage its use should the technol-ogy demonstrate some potential The limits of knowledge regard-ing effectiveness should also be explained along with what will be done to improve technological capabilities as understanding evolves
Disclosure and AuthorizationConsent
bull A clear and concise module consisting of basic disclosure and vol-untary authorization should be developed to accompany DCTT This module should not take the form of ldquoclickwraprdquo terms of service or end-user agreements but rather provide only essential information necessary for an individual to make a decision More detailed disclosures (such as FAQs in plain language) should be made easily accessible to those who wish to learn more with no hidden surprises
bull An opt-in approach to authorization should be instituted to ac-company initial DCTT rollout The feasibility and value of opt-out approaches should continue to be evaluated informed by what is technologically possible what local assessments of benefits and harms of the technology reveal over time and our evolving un-derstanding of the degree to which an opt-out approach is likely to increase or decrease utilization among different populations Opt-out approaches should not be precluded
Promoting Equity and Fairness
bull A commitment to equity means a commitment to ensuring that the benefits and burdens of DCTT are distributed fairly Public engagement is an important tool for assessing impact and to rectify inequities
bull States localities and institutions that recommend widespread use of DCTT should provide technology (eg mobile phones Blue-tooth devices) and free data packages to those who desire but lack access to these devices
Recommendations 101
bull If there are lower rates of adoption of DCTT systems in some identifiable communities public health authorities should iden-tify ways to compensate For example directing more non-DCTT resources and efforts toward those communities to meet specific needs that are elsewhere being supported by technology
bull If maps are generated based on DCTT to provide the public with the locations that COV+ individuals have visited steps must be taken to minimize the stigma and potential financial losses that could result from being identified as a hotspot
Instituting Transparent Governance and Oversight
bull Digital surveillance oversight committees should be established expeditiously with diverse and qualified membership to provide ethical and regulatory review prior to and concurrent with wide-spread use of a DCTT system
bull Understandable and publicly accessible rules must guide the col-lection access control use storage and combination of data by government authorities public and private institutions and other parties such as public health researchers
bull Only those data that are necessary and relevant for the public health response to COVID-19 should be collected and used
bull Identifiable data should be kept only for the period of time needed for the public health response to COVID-19
bull Identifiable data collected as part of this response should not be shared with anyone other than the relevant public health authorities without additional specific informed consent of individual users
bull Before a government or institution adopts a digital contact tracing program they should state the conditions under which the digital contact tracing program will be terminated
bull Future use of DCTT to advance public health or other efforts (eg use in seasonal flu surveillance) would require independent justification DCTT designed for public health use should not be used by law or immigration enforcement
102 Digital Contact Tracing for Pandemic Response
bull The principles offered in this guidance document apply both during and following the COVID-19 pandemic
Legislative
bull Congress should enact new legislation specifically tailored to fa-cilitate the use of DCTT as part of the public health response to COVID-19 while also protecting user privacy and ensuring data security
bull Congress should require DCTT developers to disclose to users in clear language the nature of the information that would be col-lected how it would be collected how it would be stored and for what purposes it may be used
bull While the rollout of DCTT should initially employ an opt-in au-thorization approach the feasibility acceptability and value of opt-out approaches should continue to be evaluated As such opt-out approaches to consent should not be precluded by legislation
bull Congress should prohibit the commercial use of data collected for COVID-19 response by DCTT
bull Congress should prohibit discrimination on the basis of data col-lected by DCTT
bull If Congress is unable to enact suitable legislation state legislatures should work toward enacting similar laws for their jurisdictions A ldquomodelrdquo state law should be rapidly developed to facilitate na-tionwide uniformity of legal requirements
Resources
103
US Government Response
White House
The White House and CDC 16 April 2020 ldquoOpening Up America Againrdquo The White House and the Centers for Disease Control and Prevention Available at httpswwwwhitehousegovopeningamerica
Congress
A Bill to Protect the Privacy of Consumersrsquo Personal Health Information
Proximity Data Device Data and Geolocation Data during the Corona-
virus Public Health Crisis S3663 116th Cong (2019-2020) Available at httpswwwcongressgovbill116th-congresssenate-bill3663
A Bill to Protect the Privacy of Health Information during a National
Health Emergency S3749 116th Cong (2019-2020) Available at httpswwwcongressgovbill116th-congresssenate-bill3749
To Protect the Privacy of Health Information during a National Health
Emergency HR6866 116th Cong (2019-2020) Available at httpswwwcongressgovbill116th-congresshouse-bill6866
Congressional Research Service
Foster Michael 16 April 2020 ldquoCOVID-19 Digital Surveillance and Privacy Fourth Amendment Considerationsrdquo Legal Sidebar LSB10449 Congres-sional Research Service httpscrsreportscongressgovproductpdfLSBLSB10449
Mulligan Stephen P and Chris D Linebaugh 25 March 2019 ldquoData Pro-tection Law An Overviewrdquo R45631 Congressional Research Service httpscrsreportscongressgovproductpdfRR45631
104 Resources
US Department of Health and Human Services (HHS) Centers for Disease Control and Prevention (CDC)
CDC 6 May 2020 ldquoInterim Guidance for Businesses and Employersrdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention Available at httpswwwcdcgovcoronavirus2019-ncov communityguidance-business-responsehtml
CDC 3 May 2020 ldquoDiscontinuation of Isolation for Persons with COVID-19 Not in Healthcare Settingsrdquo Interim Guidance Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention httpswwwcdc govcoronavirus2019-ncovhcpdisposition-in-home-patientshtml
CDC 30 April 2020 ldquoContact Tracingrdquo Get and Keep America Open Sup-porting States Tribes Localities and Territories Centers for Disease Control and Prevention httpswwwcdcgovcoronavirus2019-ncovphpopen-americacontact-tracinghtml
CDC 29 April 2020 ldquoInterim Guidelines for Collecting Handling and Testing Clinical Specimens from Persons for Coronavirus Disease 2019 (COVID-19)rdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention Available at httpswwwcdcgovcorona-virus2019-ncovlabguidelines-clinical-specimenshtml
CDC 29 April 2020 ldquoContact Tracing Part of a Multipronged Approach to Fight the COVID-19 Pandemicrdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention httpswwwcdcgov coronavirus2019-ncovphpprinciples-contact-tracinghtml
CDC 28 April 2020 ldquoPreliminary Criteria for the Evaluation of Digital Contact Tracing Tools for COVID-19rdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention Available at httpswwwcdc govcoronavirus2019-ncovdownloadsphpprelim-eval-criteria-digital- contact-tracingpdf
CDC 20 April 2020 ldquoDigital Contact Tracing Tools for COVID-19rdquo Corona- virus Disease 2019 (COVID-19) Centers for Disease Control and Preven-tion Available at httpswwwcdcgovcoronavirus2019-ncovdownloadsdigital-contact-tracingpdf
CDC 6 April 2020 ldquoOperational Considerations for the Identification of Healthcare Workers and Inpatients with Suspected COVID-19 in Non-US Healthcare Settingsrdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention Available at httpswwwcdcgovcoron virus2019-ncovhcpnon-us-settingsguidance-identify-hcw-patientshtml
HHS 2 April 2020 ldquoNotification of Enforcement Discretion under HIPAA to Allow Uses and Disclosures of Protected Health Information by Business
Resources 105
Associates for Public Health and Health Oversight Activities in Response to COVID-19rdquo 45 CFR Parts 160 and 164 Available at httpswwwhhsgovsitesdefaultfilesnotification-enforcement-discretion-hipaapdf
HHS 30 March 2020 Notification of Enforcement Discretion for Telehealth Remote Communications during the COVID-19 Nationwide Public Health Emergency US Department of Health and Human Services Office of Civil Rights Available at httpswwwhhsgovhipaafor-professionalsspecial- topicsemergency-preparednessnotification-enforcement-discretion- telehealthindexhtml
CDC 24 February 2020 ldquoLegal Authorities for Isolation and Quarantinerdquo Centers for Disease Control and Prevention Available at httpswww cdcgovquarantineaboutlawsregulationsquarantineisolationhtml
HHS 26 November 2019 ldquoPublic Health Emergency Declarationrdquo Public Health Emergency US Department of Health and Human Services Office of the Assistant Secretary for Preparedness and Response Available at httpswwwphegovPreparednesslegalPagesphedeclarationaspx
CDC September 2018 ldquoPublic Health Surveillance Preparing for the Futurerdquo Centers for Disease Control and Prevention Available at httpswww cdcgovsurveillancepdfsSurveillance-Series-Booklethpdf
CDC 2005 ldquoVI Plan for Surveillance of Contacts of SARS Case Supplement B SARS Surveillancerdquo Centers for Disease Control and Prevention Available at httpswwwcdcgovsarsguidanceb-surveillancecontactshtml
National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research 1979 ldquoThe Belmont Report Ethical Principles and Guidelines for the Protection of Human Subjects of Researchrdquo US Depart-ment of Health and Human Services Office of Human Research Protec-tions httpswwwhhsgovohrpregulations-and-policybelmont-reportread-the-belmont-reportindexhtml
US Equal Employment Opportunity Commission (EEOC)
EEOC 7 May 2020 ldquoWhat You Should Know About COVID-19 and the ADA the Rehabilitation Act and Other EEO Lawsrdquo US Equal Employ-ment Opportunity Commission Available at httpswwweeocgovwyskwhat-you-should-know-about-covid-19-and-ada-rehabilitation-act-and-other-eeo-laws
EEOC 21 March 2020 ldquoPandemic Preparedness in the Workplace and the Americans with Disabilities Actrdquo US Equal Employment Oppor-tunity Commission Available at httpswwweeocgovlawsguidancepandemic-preparedness-workplace-and-americans-disabilities-act
106 Resources
Federal Trade Commission (FTC)
FTC 2020 ldquoPrivacy amp Data Security Update 2019rdquo Federal Trade Commis-sion httpswwwftcgovsystemfilesdocumentsreportsprivacy-data- security-update-20192019-privacy-data-security-report-508pdf
FTC 2012 ldquoProtecting Consumer Privacy in an Era of Rapid Changerdquo FTC Report Recommendations for Businesses and Policymakers Fed-eral Trade Commission httpswwwftcgovsitesdefaultfilesdocumentsreportsfederal-trade-commission-report-protecting-consumer-privacy-era- rapid-change-recommendations120326privacyreportpdf
Occupational Safety and Health Administration (OSHA)
OSHA April 2020 ldquoGuidance on Preparing Workplaces for COVID-19rdquo OSHA 3990-03 2020 Department of Labor Occupational Safety and Health Act of 1970 Available at httpswwwoshagovPublicationsOSHA3990pdf
Customs and Border Protection (CBP)
CBP 4 January 2018 ldquoBorder Search of Electronic Mediardquo US Customs and Border Protection CDP DIRECTIVE NO 3340-049A Available at httpswwwcbpgovsitesdefaultfilesassetsdocuments2018-JanCBP-Directive-3340-049A-Border-Search-of-Electronic-Media-Compliantpdf
Other Governmental and Nongovernmental Organizations
Contact Tracing Surveillance PlansMethods
Simmons-Duffin Selena 7 May 2020 ldquoStates Nearly Doubled Plans for Contact Tracers since NPR Surveyed Them 10 Days Agordquo NPR Available at httpswwwnprorgsectionshealth-shots20200428846736937we-asked-all-50-states-about-their-contact-tracing-capacity-heres-what- we-learned
Resolve to Save Lives 2020 ldquoCOVID-19 Contact Tracing Playbookrdquo Vital Strategies Available at httpscontacttracingplaybookresolvetosavelives org
Simpson Erin and Adam Conner 22 April 2020 ldquoDigital Contact Tracing to Contain the Coronavirusrdquo Center for American Progress Avail-able at httpswwwamericanprogressorgissuestechnology-policynews20200422483521digital-contact-tracing-contain-coronavirus
Watson Crystal Anita Cicero James Blumenstock and Michael Fraser 10 April 2020 ldquoA National Plan to Enable Comprehensive COVID-19 Case Finding and Contact Tracing in the USrdquo Johns Hopkins Bloomberg School of Pub-
Resources 107
lic Health Center for Health Security and the Association of State and Ter-ritorial Health Officials Available at httpswwwcenterforhealthsecurity orgour-workpubs_archivepubs-pdfs2020200410-national-plan-to- contact-tracingpdf
Africa CDC 9 April 2020 ldquoGuidance on Contact Tracing for COVID-19 Pandemicrdquo Manuals Guidelines amp Frameworks African Union Africa CDC Available at httpsafricacdcorgdownloadguidance-on-contact- tracing-for-covid-19-pandemic
PIH 4 April 2020 ldquoPart I Testing Contact Tracing and Community Manage-ment of COVID-19rdquo PIH Guide | COVID-19 STOP COVID Partners in Health Available at httpswwwpihorgsitesdefaultfiles2020-04PIH_Guide_COVID_Part_I_Testing_Tracing_Community_Managment_4_4pdf
Hellewell Joel Sam Abbott Amy Gimma et al 28 February 2020 ldquoFea- sibility of Controlling COVID-19 Outbreaks by Isolation of Cases and Contactsrdquo The Lancet Global Health 8 e488ndash96 Available at https doiorg101016S2214-109X(20)30074-7
Contact Tracing EthicsPrinciplesGuidance
Center for Democracy amp Technology (CDT)ndashhttpscdtorginsightsCDT 30 April 2020 ldquoStatement of the Center for Democracy amp Technol-
ogy Regarding Use of Data to Fight COVID-19rdquo Center for Democ-racy amp Technology httpscdtorgwp-contentuploads202004 CDT-Statement-Regarding-Use-of-Data-to-Fight-COVID-19pdf
Shetty Ridhi 23 April 2020 ldquoData Use in the Fight against COVID-19 Should Treat People Equitably Not Exacerbate Long-Standing Dispar-itiesrdquo Center for Democracy amp Technology httpscdtorginsightsdata-use-in-the-fight-against-covid-19-should-treat-people-equitably-not-exacerbate-long-standing-disparities
Greenwood Dazza Gregory Nadeau Pagona Tsormpatzoudi Bryan Wilson Jeffrey Saviano and Alex ldquoSandyrdquo Pentland 30 April 2020 ldquoCOVID-19 Contact Tracing Privacy Principlesrdquo MIT Computational Law Report Available at httpslawmitedupubcovid19contacttracingprivacy- principlesrelease7
Editorial Board 29 April 2020 ldquoShow Evidence That Apps for COVID-19 Contact-Tracing Are Secure and Effectiverdquo Nature 580 (7805) 563ndash563 httpsdoiorg101038d41586-020-01264-1
ldquoJoint Statement on Contact Tracingrdquo 19 April 2020 Available at httpscryptobriefingcomwp-contentuploads202004Joint-State-ment-from-Researcherspdf
Hinch Robert Will Probert Anel Nurtay Michelle Kendall Chris Wymant
108 Resources
Matthew Hall Katrina Lythgoe et al 16 April 2020 ldquoEffective Configura-tions of a Digital Contact Tracing App A Report to NHSXrdquo Available at httpsgithubcomBDI-pathogenscovid-19_instant_tracingblobmasterReportndashEffective Configurations of a Digital Contact Tracing Apppdf
Kahn Gilmor Daniel 16 April 2020 ldquoPrinciples for Technology-Assisted Contact-Tracingrdquo White Paper American Civil Liberties Union httpswwwacluorgreportaclu-white-paper-principles-technology-assisted- contact-tracing
Carroll Anna and Samantha Stroman 16 April 2020 ldquoFind My Friends in a Pandemic The Future of Contact Tracing in Americardquo CSIS Commis-sion on Strengthening Americarsquos Health Security April 16 2020 httpshealthsecuritycsisorgarticlesfind-my-friends-in-a-pandemic-the-future- of-contact-tracing-in-america
Electronic Privacy Information Center Testimony to Congress 15 April 2020 ldquoEPIC to Congress Establish Privacy Safeguards for Digital Contact Tracingrdquo httpsepicorgtestimonycongressEPIC-HEC-Contact-Tracing-Apr2020pdf
Crocker Andrew Kurt Opsahl and Bennett Cyphers 10 April 2020 ldquoThe Chal-lenge of Proximity Apps for COVID-19 Contact Tracingrdquo Electronic Fron-tier Foundation April 10 2020 httpswwwefforgdeeplinks202004challenge-proximity-apps-covid-19-contact-tracing
Hart Vi Divya Siddarth Bethan Cantrell Lila Tretikov Peter Eckersley John Langford Scott Leibrand et al 3 April 2020 ldquoOutpacing the Virus Digital Response to Containing the Spread of COVID-19 While Mitigat-ing Privacy Risksrdquo Whitepaper 5 COVID-19 Rapid Response Impact Initiative Edmond J Safra Center for Ethics httpsdrivegooglecomfiled1vIN2AX-DDNW-S0aHq8xs0RJ2jkR_CckXview
Raskar Ramesh Isabel Schunemann Rachel Barbar Kristen Vilcans Jim Gray Praneeth Vepakomma Suraj Kapa Andrea Nuzzo Rajiv Gupta et al 19 March 2020 ldquoApps Gone Rogue Maintaining Personal Privacy in an Epidemicrdquo White Paper Private Kit MIT httpsarxivorgpdf2003 08567pdf
General (not COVID-19 specific) Statements of Principles Legal Frameworks Other Information
Nuffield Council on Bioethics 2020 ldquoGuide to the Ethics of Surveillance and Quarantine for Novel Coronavirusrdquo httpswwwnuffieldbioethicsorgassetspdfsGuide-to-the-ethics-of-surveillance-and-quarantine-for-novel-coronaviruspdf
Resources 109
Schwartz Adam 2020 ldquoHow EFF Evaluates Government Demands for New Surveillance Powersrdquo Electronic Frontier Foundation April 3 2020 httpswwwefforgdeeplinks202004how-eff-evaluates-government-demands- new-surveillance-powers
WHO 2 April 2020 ldquoCoronavirus Disease 2019 (COVID-19)rdquo Situation Report 73 World Health Organization httpswwwwhointdocsdefault-sourcecoronavirusesituation-reports20200402-sitrep-73-covid-19pdf
Nuffield Council on Bioethics 17 March 2020 ldquoEthical Considerations in Responding to the COVID-19 Pandemicrdquo Rapid Policy Briefing Nuffield Council on Bioethics httpswwwnuffieldbioethicsorgassetspdfs Ethical-considerations-in-responding-to-the-COVID-19-pandemicpdf
Nuffield Council on Bioethics 28 January 2020 ldquoResearch in Global Health Emergencies Ethical Issuesrdquo httpswwwnuffieldbioethicsorgassetspdfsRGHE_full_report1pdf
Vota Wayan 4 December 2019 ldquoEvery African Countryrsquos National eHealth Strategy or Digital Health Policyrdquo ICT Works httpswwwictworksorgafrican-national-ehealth-strategy-policy
ENISA 3 December 2019 ldquoPseudonymisation Techniques and Best Practicesrdquo European Union Agency for Cybersecurity httpswwwenisaeuropaeupublicationspseudonymisation-techniques-and-best-practices
ldquoMobile Cellular Subscriptions (per 100 People)rdquo 2018 World Telecommuni-cationICT Development Report International Telecommunication Union httpsdataworldbankorgindicatorITCELSETSP2
Levinson-Waldman Rachel 2018 ldquoCellphones Law Enforcement and the Right to Privacyrdquo Brennan Center for Justice New York University School of Law httpswwwbrennancenterorgsitesdefaultfiles2019-08Report_Cell_Surveillance_Privacypdf
Code of Federal Regulations Part 46ndashProtection of Human Subjects (45 CFR 46 US Department of Health and Human Services)
General Data Protection Regulation (REGULATION (EU) 2016679 (GDPR) European Union)
European Data Protection Supervisor ldquoNecessity and Proportionalityrdquo httpsedpseuropaeudata-protectionour-worksubjectsnecessity- proportionality_en
ISOIEC 38505-12017 Information technologymdashGovernance of ITmdashGover-nance of datamdashPart 1 Application of ISOIEC 38500 to the governance of data (2017 International Organization for Standardization (ISO))
WHO 2016 ldquoWHO Guidance for Managing Ethical Issues in Infectious Dis-ease Outbreaksrdquo World Health Organization httpswwwwhointethicspublicationsinfectious-disease-outbreaksen
110 Resources
CIOMS 2016 ldquoInternational Ethical Guidelines for Health-Related Research Involving HumansndashGuideline 22 Use of Data Obtained from the Online Environment and Digital Tools in Health Related Researchrdquo Council for International Organizations of Medical Sciences httpsciomschwp- contentuploads201701WEB-CIOMS-EthicalGuidelinespdf
Principles for Digital Development 2016 ldquoPrinciplesrdquo httpsdigitalprinciplesorgprinciples
GA4GH 2014 ldquoFramework for Responsible Sharing of Genomic and Health-Related Datardquo Global Alliance for Genomics amp Health 9 Decem-ber 2014 httpswwwga4ghorggenomic-data-toolkitregulatory-ethics- toolkitframework-for-responsible-sharing-of-genomic-and-health-related- data
WHO and PATH 2013 ldquoPlanning an Information Systems Project A Toolkit for Public Health Managersrdquo World Health Organization amp PATH httpspathazureedgenetmediadocumentsTS_opt_ict_toolkitpdf
ASTHO 2012 ldquoPublic Health Collection Use Sharing and Protection of Informationrdquo Issue Brief ASTHO Legal Preparedness Series Association of State and Territorial Health Officials httpswwwasthoorgProgramsPreparednessPublic-Health-Emergency-LawPublic-Health-and-Informat- ion-Sharing-ToolkitCollection-Use-Sharing-and-Protection-Issue-Brief
Cavoukian Ann 2010 ldquoPrivacy by Design The 7 Foundational Principlesrdquo Ontario Canada Information and Privacy Commissioner of Canada httpsiapporgmediapdfresource_centerpbd_implement_7found_ principlespdf
PHLS 2002 ldquoPrinciples of the Ethical Practice of Public Healthrdquo Public Health Leadership Society httpswwwaphaorg-mediafilespdfmembergroupsethicsethics_brochureashx
National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research 1979 ldquoThe Belmont Report Ethical Principles and Guidelines for the Protection of Human Subjects of Researchrdquo US Dept of Health amp Human Services Office of Human Research Protections httpswwwhhsgovohrpregulations-and-policybelmont-reportread-the-bel mont-reportindexhtml
Reopening ProposalsPlans (Broad)
United Kingdom Cabinet Office 2020 ldquoOur Plan to Rebuild The UK Govern-mentrsquos COVID-19 Recovery Strategyrdquo 12 May 2020 httpswwwgovukgovernmentpublicationsour-plan-to-rebuild-the-uk-governments-covid- 19-recovery-strategyour-plan-to-rebuild-the-uk-governments-covid-19- recovery-strategyfourteen-supporting-programmes
Resources 111
Shannon Joel Lorenzo Reyes and Doyle Rice 2020 ldquoAre Lockdowns Being Relaxed in My State Herersquos How America Is Reopening amid the Coronavirus Pandemicrdquo USA TODAY May 21 2020 httpswww usatodaycomstorynewshealth20200419coronavirus-lockdown- reopening-states-us-texas-florida5155269002
Romer Paul 2020 ldquoRoadmap to Responsibly Reopen Americardquo 23 April 2020 httpsroadmappaulromernetpaulromer-roadmap-reportpdf
Governor Larry Hogan 24 April 2020 ldquoMaryland Strong Roadmap to Recoveryrdquo httpsgovernormarylandgovwp-contentuploads202004MD_Strongpdf
NGA and ASTHO 21 April 2020 ldquoRoadmap to Recovery A Public Health Guide for Governorsrdquo National Governors Association and American State and Territorial Health Officials httpswwwngaorgwp-contentuploads202004NGA-Reportpdf
Allen Danielle Julius Krein Ganesh Sitaraman and E Glen Weyl 2020 ldquoNational Covid-19 Testing Action Planrdquo The Rockefeller Foundation httpswwwrockefellerfoundationorgwp-contentuploads202004TheRockefellerFoundation_WhitePaper_Covid19_4_22_2020pdf
Edmond J Safra Center for Ethics Harvard UniverityndashWhite PapersAllen Danielle Sharon Block Joshua Cohen et al 20 April 2020 ldquoRoad-
map to Pandemic Resiliencerdquo httpsethicsharvardedufilescenter- for-ethicsfilesroadmaptopandemicresilience_updated_42020_0pdf
Allen Danielle Lucas Stanczyk Rajiv Sethi Glen Weyl 25 March 2020 ldquoWhen Can We Go Outrdquo httpsdrivegooglecomfiled1gf21eYeNWwrR9OO5nzxn1jlv-RTmHkt0view
Mulheirn Ian Sam Alvis Lizzie Insall James Browne Christina Palmou 20 April 2020 ldquoA Sustainable Exit Strategy Managing Uncertainty Minimis-ing Harmrdquo Tony Blair Institute for Global Change httpsinstituteglobalsitesdefaultfilesinline-filesA20Sustainable20Exit20Strategy2C 20Managing20Uncertainty20Minimising20Harmpdf
Emanuel Zeke Neera Tanden Adam Conner Erin Simpson Nicole Rap-fogel and Maura Calsyn 2020 ldquoA National and State Plan to End the Coronavirus Crisisrdquo Center for American Progress April 3 2020 httpswwwamericanprogressorgissueshealthcarenews20200403482613national-state-plan-end-coronavirus-crisis
Gottlieb Scott Caitlin Rivers Mark McClellan Lauren Silvis and Crystal Watson 2020 ldquoNational Coronavirus Response A Road Map to Reopen-ingrdquo American Enterprise Institute httpswwwaeiorgresearch-productsreportnational-coronavirus-response-a-road-map-to-reopening
112 Resources
Digital Contact Tracing Experiences from Other Countries
Multiple
Woodhams Samuel 12 May 2020 ldquoCOVID-19 Digital Rights Trackerrdquo Top10VPN httpswwwtop10vpncomresearchinvestigationscovid-19- digital-rights-tracker
Fahim Kareem Min Joo Kim and Steve Hendrix 2 May 2020 ldquoCellphone Monitoring Is Spreading with the Coronavirus So Is an Uneasy Tolerance of Surveillancerdquo Washington Post (Washington DC) httpswww washingtonpostcomworldcellphone-monitoring-is-spreading-with-the- coronavirus-so-is-an-uneasy-tolerance-of-surveillance20200502 56f14466-7b55-11ea-a311-adb1344719a9_storyhtml
Ikram Umar Christer Mjaringset MD Anne-Marie Boxall Mylaine Breton Ines Gravey Holly Krelle Veacuteronique Raimond and Reginald D Williams II 30 April 2020 ldquoWhat Can the US Learn from Innovative Strategies Used in Other Countries to Respond to COVID-19rdquo The Commonwealth Fund httpswwwcommonwealthfundorgblog2020what-can-us-learn-innovative-strategies-used-other-countries-respond-covid-19
Jens-Henrik Jeppesen and Pasquale Esposito 29 April 2020 ldquoCOVID-19 European Data Collection and Contact Tracing Measuresrdquo Center for Democracy amp Technology httpscdtorginsightscovid-19-european- data-collection-and-contact-tracing-measures
Africa CDC 9 April 2020 ldquoGuidance on Contact Tracing for COVID-19 Pan-demicrdquo Manuals Guidelines amp Frameworks African Union Africa CDC httpsafricacdcorgdownloadguidance-on-contact-tracing-for-covid-19- pandemic
Heneghan Carl Jon Brassey and Tom Jefferson 6 April 2020 ldquoCOVID-19 What Proportion Are Asymptomaticrdquo Centre for Evidence-Based Medi-cine httpswwwcebmnetcovid-19covid-19-what-proportion-are- asymptomatic
Kharpal Arjun 30 March 2020 ldquoUse of Surveillance to Fight Coronavirus Raises Concerns about Government Power after Pandemic Endsrdquo CNBC httpswwwcnbccom20200327coronavirus-surveillance-used-by- governments-to-fight-pandemic-privacy-concernshtml
Canada
ldquoCommissioner Publishes Framework to Assess Privacy-Impactful Initiatives in Response to COVID19rdquo Office of the Privacy Commissioner of Canada 17 April 2020 httpswwwprivgccaenopc-newsnews-and-announcements 2020an_200417
Resources 113
China
Kraemer Moritz U G Chia-Hung Yang Bernardo Gutierrez Chieh-Hsi Wu Brennan Klein David M Pigott Open COVID-19 Data Working Group Louis du Plessis Nuno R Faria Ruoran Li William P Hanage John S Brownstein Maylis Layan Alessandro Vespignani Huaiyu Tian Chris-topher Dye Oliver G Pybus Samuel V Scarpino ldquoThe Effect of Human Mobility and Control Measures on the COVID-19 Epidemic in Chinardquo Science 368(6490) 493-497 DOI 101126scienceabb4218
Bi Qifang Yongsheng Wu Shujiang Mei Chenfei Ye Xuan Zou Zhen Zhang Xiaojian Liu Lan Wei Shaun A Truelove Tong Zhang Wei Gao Cong Cheng Xiujuan Tang Xiaoliang Wu Yu Wu Binbin Sun Suli Huang Yu Sun Juncen Zhang Ting Ma Justin Lessler and Teijian Feng ldquoEpidemiol-ogy and Transmission of COVID-19 in 391 Cases and 1286 of Their Close Contacts in Shenzhen China A Retrospective Cohort Studyrdquo The Lancet April 27 2020 DOIhttpsdoiorg101016S1473-3099(20)30287-5
Sun Kaiyuan and Ceacutecile Viboud ldquoImpact of contact tracing on SARS-CoV-2 transmissionrdquo The Lancet April 27 2020 DOIhttpsdoiorg101016S1473-3099(20)30357-1
Ferretti Luca Chris Wymant Michelle Kendall Lele Zhao Anel Nurtay Lucie Abeler-Doumlrner Michael Parker David Bonsall and Christophe Fraser 2020 ldquoQuantifying SARS-CoV-2 Transmission Suggests Epidemic Control with Digital Contact Tracingrdquo Science 368 (6491) httpsdoiorg101126scienceabb6936
Gan Nectar and David Culver 2020 ldquoChina Is Fighting the Coronavirus with a Digital QR code Herersquos How It Worksrdquo CNN Business April 16 2020 httpswwwcnncom20200415asiachina-coronavirus-qr-code-intl-hnkindexhtml
Mozur Paul Raymond Zhong and Aaron Krolik 2020 ldquoIn Coronavirus Fight China Gives Citizens a Color Code with Red Flagsrdquo New York Times (New York NY) March 1 2020 httpswwwnytimescom20200301businesschina-coronavirus-surveillancehtml
Germany
Schwartz Matthew S 2020 ldquoGermany Backs Away from Compiling Coro-navirus Contacts in a Central Databaserdquo NPR April 27 2020 https wwwnprorgsectionscoronavirus-live-updates20200427846046185 germany-backs-away-from-compiling-coronavirus-contacts-in-a-central- databaseutm_medium=RSSamputm_campaign=news
Busvine Douglas and Andreas Rinke 2020 ldquoGermany Flips to Apple-Google
114 Resources
Approach on Smartphone Contact Tracingrdquo Reuters April 26 2020 httpswwwreuterscomarticleus-health-coronavirus-europe-techgermany-flips-on-smartphone-contact-tracing-backs-apple-and-google-idUSKCN22807J
Hong Kong
ldquo lsquoStayHomeSafersquo Mobile App User Guiderdquo The Government of the Hong Kong
Special Administrative Region May 20 2020httpswwwcoronavirusgov hkengstay-home-safehtml
India
Dixit Pranav 2020 ldquoIndiarsquos Contact Tracing App Is All But Mandatory So This Programmer Hacked It So That He Always Appears Saferdquo BuzzFeed News May 12 2020 httpswwwbuzzfeednewscomarticlepranavdixitindia-aarogya-setu-hacked
Greenberg Andy 2020 ldquoIndiarsquos Covid-19 Contact Tracing App Could Leak Patient Locationsrdquo WIRED May 6 2020 httpswwwwiredcomstoryindia-covid-19-contract-tracing-app-patient-location-privacy
Alderson Elliot ldquoAarogya Setu The Story of a Failurerdquo Medium May 6 2020 httpsmediumcomfs0c131yaarogya-setu-the-story-of-a-failure- 3a190a18e34
OrsquoNeill Patrick Howell ldquoIndia Is Forcing People to Use Its COVID App Unlike Any Other Democracyrdquo MIT Technology Review May 6 2020 httpswwwtechnologyreviewcom202005071001360india-aarogya- setu-covid-app-mandatory
Israel
Hendrix Steve and Ruth Eglash 2020 ldquoIsrael Is Using Cellphone Surveil-lance to Warn Citizens You May Already Be Infectedrdquo Washington Post (Washington DC) March 19 2020 httpswwwwashingtonpostcomworldmiddle_eastisrael-is-using-cellphone-surveillance-to-warn-citizens-you-may-already-be-infected2020031968267294-69e7-11ea-b199-3a9799c54512_storyhtml
Singapore
ldquoTraceTogether Safer Togetherrdquo A Singapore Government Agency Website accessed May 21 2020 httpswwwtracetogethergovsg
Resources 115
South Korea
Korean Ministry of Health and Welfare ldquoConfirmed Patient Movement Path Websiterdquo Central Accident Remediation Headquarters accessed May 21 2020 httpncovmohwgokrbdBoardList_RealdobrdId=1ampbrdGu-bun=12ampncvContSeq=ampcontSeq=ampboard_id=ampgubun=
Kim Max S 2020 ldquoSeoulrsquos Radical Experiment in Digital Contact Tracingrdquo The New Yorker April 17 2020 httpswwwnewyorkercomnewsnews-deskseouls-radical-experiment-in-digital-contact-tracing
Watson Ivan and Sophie Jeong 2020 ldquoCoronavirus Mobile Apps Are Surging in Popularity in South Koreardquo CNN Business February 28 2020 httpseditioncnncom20200228techkorea-coronavirus-tracking-appsindexhtml
United Kingdom
Hern Alex and Kate Proctor 2020 ldquoUK May Ditch NHS Contact-Tracing App for Apple and Google Modelrdquo The Guardian May 7 2020 httpswwwtheguardiancomtechnology2020may07uk-may-ditch-nhs-contact- tracing-app-for-apple-and-google-model
Lovejoy Ben 2020 ldquoHands-on with UKrsquos NHS Contact Tracing App as the Test Goes Liverdquo 9-5 Mac Blog May 7 2020 https9to5maccom2020 0507nhs-contact-tracing
ldquoCoronavirus Test Track and Trace Plan Launched on Isle of Wightrdquo Depart-ment of Health and Social Care Press Release May 4 2020 httpswww govukgovernmentnewscoronavirus-test-track-and-trace-plan-launched- on-isle-of-wight
Leprince-Ringuet Daphne 2020 ldquoContact-Tracing Apps Why the NHS Said No to Apple and Googlersquos Planrdquo ZDNet April 28 2020 httpswww zdnetcomarticlecontact-tracing-apps-why-the-nhs-said-no-to-apple- and-googles-plan
Specific Digital ProductsApps
COVID-19
Vota Wayan 2020 Additional Proposed Coronavirus Solutions Google Doc-ument httpsdocsgooglecomspreadsheetsd15hkhdtGNzx7oHkO8Y 2MOiY83JsHjqxL4MhMGvlA_J6Ieditgid=579623365
Apple and Google nd ldquoPrivacy-Preserving Contact Tracingrdquo Apple Accessed May 19 2020 httpswwwapplecomcovid19contacttracing
116 Resources
Woodhams Samuel 12 May 2020 ldquoCOVID-19 Digital Rights Trackerrdquo Top10VPN httpswwwtop10vpncomresearchinvestigationscovid-19-digital-rights-tracker
Starobinski David and Johannes Becker 30 April 2020 ldquoHow Apple and Goo-gle Will Let Your Phone Warn You If Yoursquove Been Exposed to the Corona-virusrdquo The Conversation Accessed May 21 2020 httptheconversation comhow-apple-and-google-will-let-your-phone-warn-you-if-youve-been-exposed-to-the-coronavirus-136597
Morrison Sara 2020 ldquoThe Apple-Google Contact Tracing Tool Gets a Beta Release and a New Risk Level Featurerdquo Vox April 24 2020 httpswwwvoxcomrecode202042421234420apple-google-contact-tracing- exposure-notification-update
Commonwealth Centre for Digital Health 9 April 2020 ldquo[Webinar] CWCDH Digital Response to COVID-19rdquo 5040 httpswwwyoutubecomwatchv=ZyE_KRWLtC8ampfeature=youtube
Johns Hopkins Medicine 23 April 2020 ldquoJohns Hopkins Medicine Remote Monitoring Program for Faculty Staff and Providers Exposed to COVID-19rdquo Johns Hopkins Medicine Occupational Health Services httpswww hopkinsmedicineorghsecovid19_emocha
MIT nd ldquoPrivate Kit Safe Paths Privacy-by-Design Covid19 Solutions Using GPS+Bluetooth for Citizens and Public Health Officialsrdquo Safepaths Accessed May 20 2020 httpssafepathsmitedu
Related Disease Detection Projects
ldquoCOVID Controlrdquo Google Play accessed May 21 2020 httpsplaygoogle comstoreappsdetailsid=jhueduJohnsHopkinsCOVIDControl
ldquoCOVID Symptom Trackerrdquo Created by Massachusetts General Hospital the Harvard TH Chan School of Public Health Kingrsquos College London and Stanford University School of Medicine accessed May 21 2020 httpscovidjoinzoecomus
ldquoSee How Your Community Is Moving around Differently Due to COVID-19rdquo Google COVID-19 Mobility Reports accessed May 21 2020 httpswww googlecomcovid19mobility
Drew David A Long H Nguyen Claire J Steves Cristina Menni Maxim Freydin Thomas Varsavsky Carole H Sudre M Jorge Cardoso Sebastien Ourselin Jonathan Wolf Tim D Spector Andrew T Chan and COPE Consortium 2020 ldquoRapid Implementation of Mobile Technology for Real-Time Epidemiology of COVID-19rdquo Science 05 May 2020 DOI 101126scienceabc0473
Resources 117
Tress Luke 2020 ldquoMaccabi Medial EarlySign Develop Algorithm to Identify High-Risk COVID-19 Casesrdquo Times of Israel April 22 2020 httpswwwtimesofisraelcommaccabi-medial-earlysign-develop-algo rithm-to-identify-high-risk-covid-19-cases
ldquoPCR Diagnostic Testing for SARS-CoV-2rdquo Center for Health Security Johns Hopkins Bloomberg School of Public Health last modified April 17 2020 httpswwwcenterforhealthsecurityorgresourcesCOVID-19COVID-19-fact-sheets200130-nCoV-diagnostics-factsheetpdf
Radin Jennifer M Nathan E Wineinger Eric J Topol and Steven R Steinhubl 2020 ldquoHarnessing Wearable Device Data to Improve State-Level Real-Time Surveillance of Influenza-Like Illness in the USA A Population-Based Studyrdquo The Lancet January 16 2020 DOIhttpsdoiorg101016S2589-7500(19)30222-5
Polling
Hargittai Eszter Minh Hao Nguyen Jaelle Fuchs Jonathan Gruber Will Marler Amanda Hunsaker and Goumlkccedile Karaoglu 2020 ldquoCovid-19 Study on Digital Media and the Coronavirus Pandemicrdquo Internet Use and Society Division Institute of Communication and Media Research University of Zurich httpwebuseorgcovid
Navigator Research 2020 ldquoPublic Opinion on Coronavirus Navigator Updaterdquo Navigating Coronavirus (blog) May 21 2020 https navigatorresearchorgnavigating-coronavirus
Russonello Giovanni 2020 ldquoBig Government For Now Most Americans Say Bring It Onrdquo The New York Times May 1 2020 sec US httpswww nytimescom20200501uspoliticscoronavirus-spending-pollshtml
ldquoWashington Post-University of Maryland National Pollrdquo 2020 Washington Post April 21 2020 httpswwwwashingtonpostcomcontext washington-post-university-of-maryland-national-poll-april-21-26-20203583b4e9-66be-4ed6-a457-f6630a550ddf
Kirzinger Ashley Liz Hamel Cailey Muntildeana Audrey Kearney and Mollyann Brodie 2020 ldquoKFF Health Tracking Poll ndash Late April 2020 Coronavirus Social Distancing and Contact Tracingrdquo Kaiser Family Foundation (blog) April 24 2020 httpswwwkfforgcoronavirus-covid-19issue-briefkff-health-tracking-poll-late-april-2020
NYC DOH 2020 ldquoAge-Adjusted Rates of Lab Confirmed COVID-19 Non-hospitalized Cases Estimated Non-Fatal Hospitalized Cases and Persons Known to Have Died per 100000 by RaceEthnicity Grouprdquo New York
118 Resources
City Department of Health httpswww1nycgovassetsdohdownloadspdfimmcovid-19-deaths-race-ethnicity-04242020-1pdf
Anderson Monica and Brooke Auxier 2020 ldquoMost Americans Donrsquot Think Cellphone Tracking Will Help Limit COVID-19 Are Divided on Whether Itrsquos Acceptablerdquo Pew Research Center (blog) April 16 2020 httpswwwpewresearchorgfact-tank20200416most-americans-dont-think-cellphone-tracking-will-help-limit-covid-19-are-divided-on-whether-its-acceptable
Daly Kyle 2020 ldquoExclusive Americans Wary of Giving up Data to Fight Coronavirusrdquo Axios April 3 2020 httpswwwaxioscomexclusive- americans-wary-of-giving-up-data-to-fight-coronavirus-330fc1d9-8b99-4a51-871b-25ee0e0591f2html
Elliott Douglas J Ana Kreacic Lorenzo Milans del Bosch and Lisa Quest nd ldquoData-Sharing in the Time of Coronavirusrdquo Oliver Wyman Forum Accessed May 19 2020 httpswwwoliverwymanforumcomfuture-of-data2020aprdata-sharing-in-the-time-of-coronavirushtml
Auxier Brooke Lee Rainie Monica Anderson Andrew Perrin Madhu Kumar and Erica Turner 2019 ldquoAmericans and Privacy Concerned Confused and Feeling Lack of Control Over Their Personal Informationrdquo Pew Research
Center Internet Science amp Tech (blog) November 15 2019 httpswww pewresearchorginternet20191115americans-and-privacy-concerned- confused-and-feeling-lack-of-control-over-their-personal-information
EPIC 2020 ldquoPublic Opinion on Privacyrdquo Electronic Privacy Information Center January 22 2020 httpsepicorgprivacysurvey
Pew Research Center 2020 ldquoDemographics of Mobile Device Ownership and Adoption in the United Statesrdquo Internet amp Technology Pew Research Cen-ter httpswwwpewresearchorginternetfact-sheetmobile
Rodrigues Rafaela Alina Husain Amanda Couture-Carron Leslye E Orloff and Nawal H Ammar 2018 ldquoPromoting Access to Justice for Immigrant and Limited English Proficient Crime Victims in an Age of Increased Immi-gration Enforcement Initial Report from a 2017 National Surveyrdquo Wash-ington DC National Immigrant Womenrsquos Advocacy Project American University Washington College of Law httpniwaplibrarywclamericaneduwp-contentuploadsImmigrant-Access-to-Justice-National-Reportpdf
Center for Survey Measurement (CSM) MEMORANDUM FOR Associate Directorate for Research and Methodology (ADRM) 2017 ldquoRespondent Confidentiality Concernsrdquo September 20 2017 httpswww2censusgovcacnacmeetings2017-11Memo-Regarding-Respondent-Confidentiality- Concernspdf
Resources 119
Pew Research Center 2017 ldquoUS Muslims Concerned About Their Place in Society but Continue to Believe in the American Dreamrdquo Religion amp Public Life Pew Research Center httpswwwpewforumorg20170726findings-from-pew-research-centers-2017-survey-of-us-muslims
Popular Press
Dixit Pranav 2020 ldquoIndiarsquos Contact Tracing App Is All But Mandatory So This Programmer Hacked It So That He Always Appears Saferdquo BuzzFeed News May 12 2020
Mills Rodrigo Chris 2020 ldquoDigital Contact Tracing Is Becoming Available but Is It Effectiverdquo The Hill May 7 2020 httpsthehillcompolicytechnology 496498-digital-contact-tracing-is-becoming-available-but-is-it-effective
Ingram David 2020 ldquoApple Google Push Makers of Coronavirus Apps Not to Record User Locationrdquo May 4 2020 NBC News Accessed May 19 2020 httpswwwnbcnewscomtechtech-newscoronavirus-apps-won-t-be-able- record-users-location-apple-n1199586
McMinn Sean 2020 ldquoMobile Phone Data Show More Americans Are Leav- ing Their Homes Despite Ordersrdquo NPR May 1 2020 httpswwwnpr org20200501849161820mobile-phone-data-show-more-americans- are-leaving-their-homes-despite-orders
Valentino-DeVries Jennifer Natasha Singer and Aaron Krolik 2020 ldquoA Scram-ble for Virus Apps That Do No Harmrdquo The New York Times April 29 2020 sec Business httpswwwnytimescom20200429 businesscoronavirus-cellphone-apps-contact-tracinghtml
Giglio Mike 2020 ldquoWould You Sacrifice Your Privacy to Get Out of Quaran-tinerdquo The Atlantic April 22 2020 httpswwwtheatlanticcompoliticsarchive202004coronavirus-pandemic-privacy-civil-liberties-911609172
Bradshaw Tim 2020 ldquo2 Billion Phones Cannot Use Google and Apple Con-tact-Tracing Tech | Ars Technicardquo Ars Technica April 20 2020 httpsarstechnicacomtech-policy2020042-billion-phones-cannot-use-google-and-apple-contract-tracing-tech
Reston Maeve Kristina Sgueglia and Cheri Mossburg 2020 ldquoGovernors on East and West Coasts Form Pacts to Decide When to Reopen Economiesrdquo CNN Politics April 13 2020 httpswwwcnncom20200413politicsstates-band-together-reopening-plansindexhtml
Cochrane Emily Claire Cain Miller and Jim Tankersley 2020 ldquoTrump Admin-istration Scales Back Paid Leave in Coronavirus Relief Lawrdquo The New York Times April 2 2020 sec US httpswwwnytimescom20200402uspoliticscoronavirus-paid-leavehtml
120 Resources
Kim Nemo 2020 ldquolsquoMore Scary than Coronavirusrsquo South Korearsquos Health Alerts Expose Private Livesrdquo The Guardian March 6 2020 sec World news httpswwwtheguardiancomworld2020mar06more-scary- than-coronavirus-south-koreas-health-alerts-expose-private-lives
Hickey Matt 2014 ldquoCarriers Can Now Install Apps On Android Handsets Without Customersrsquo Permissionrdquo Forbes December 1 2014 httpswww forbescomsitesmatthickey20141201carriers-can-now-install-apps-on- android-handsets-without-customers-permission
Commentaries
Giubilini Alberto 2020 ldquoContact-Tracing Apps and the Future COVID-19 Vaccination Should Be Compulsory Social Technological and Pharmaco-logical Immunisationrdquo Practical Ethics (blog) May 6 2020 httpblo gpracticalethicsoxacuk202005contact-tracing-apps-and-the-future- covid-19-vaccination-should-be-compulsory-social-technological-and- pharmacological-immunisation
Landau Susan Christy Lopez and Laura Moy 2020 ldquoThe Importance of Equity in Contact Tracingrdquo Lawfare (blog) May 1 2020 httpswww lawfareblogcomimportance-equity-contact-tracing
Schaefer G Owen and Angela Ballantyne 2020 ldquoDownloading COVID-19 Contact Tracing Apps Is a Moral Obligationrdquo Journal of Medical Ethics
Blog (blog) May 4 2020 httpsblogsbmjcommedical-ethics20200504downloading-covid-19-contact-tracing-apps-is-a-moral-obligation
OrsquoNeill Patrick Howell 2020 ldquoFive Things We Need to Do to Make Contact Tracing Really Workrdquo MIT Technology Review April 28 2020 httpswwwtechnologyreviewcom202004281000714five-things-to-make- contact-tracing-work-covid-pandemic-apple-google
Doffman Zak 2020 ldquoCOVID-19 Contact Tracing Why Apple And Google Canrsquot Make This Workrdquo Forbes April 27 2020 httpswwwforbescomsiteszakdoffman20200427this-is-the-contact-tracing-worry-even-apple- and-google-cant-resolve
All Tech is Human 2020 The Ethics of Contact Tracing for COVID-19 httpswwwyoutubecomwatchv=59mKUAVDhdkampt=626s
Canca Cansu 2020 ldquoWhy lsquoMandatory Privacy-Preserving Digital Contact Tracingrsquo Is the Ethical Measure against COVID-19rdquo Medium May 4 2020 httpsmediumcomcansucancawhy-mandatory-privacy-preserv-ing-digital-contact-tracing-is-the-ethical-measure-against-covid-19-a0d-143b7c3b6
Goodman Bryce 2020 ldquoCOVID and Contact Tracing When Social Justice
Resources 121
Demands Mass Surveillancerdquo Medium April 10 2020 httpsmediumcombwgoodmancovid-and-contact-tracing-when-social-justice-demands-mass-surveillance-18d419b8cc5
Gray Rosie and Caroline Haskins 2020 ldquoThey Were Opposed To Govern-ment Surveillance Then The Coronavirus Pandemic Beganrdquo BuzzFeed News March 30 2020 httpswwwbuzzfeednewscomarticlerosiegraythey-were-opposed-to-government-surveillance-then-the
Cegłowski Maciej 2020 ldquoWe Need A Massive Surveillance Programrdquo Idle
Words (blog) March 23 2020 httpsidlewordscom202003we_need_ a_massive_surveillance_programhtm
Academic Literature
COVID-19 Specific
Abeler J Baumlcker M Buermeyer U Zillessen H (2020) COVID-19 Contact Tracing and Data Protection Can Go Together JMIR mHealth and uHealth 8(4) e19359 doi 10219619359
Altmann S Milsom L Zillessen H et al (2020) Acceptability of App-Based Contact Tracing for COVID-19 Cross-Country Survey Evidence Preprint
Berke A Bakker M Vepakomma P Larson K Pentland A (2020) Assessing Disease Exposure Risk with Location Data A Proposal for Cryptographic Preservation of Privacy arXiv arXiv200314412ndashMarch 2020
Bonsail D Parker M Fraser C (2020) Sustainable Containment of COVID-19 Using Smartphones in China Scientific and Ethical Underpinnings for Implementation of Similar Approaches in Other Settings Unpublished working paper
Bradshaw WJ Alley EC Huggins JH Lloyd AL Esvelt KM (2020) Bidirectional Contact Tracing Is Required for Reliable COVID-19 Control Preprint via MedRxiv
Braithwaite I Callender T Bullock M Aldridge R (2020) Automated and Semi-Automated Contact Tracing Protocol for a Rapid Review of Avail-able Evidence and Current Challenges to Inform the Control of COVID-19 Preprint via medRxiv doi httpsdoiorg1011012020041420063636
Bulchandani Bannerjee V Shivam S Moudgalya S Sondhi SL (2020) Digital Herd Immunity and COVID-19 Preprint via medRxiv doi httpsdoiorg1011012020041520066720
Cheng H Jian S Liu D (2020) Contact Tracing Assessment of COVID-19 Transmission Dynamics in Taiwan and Risk at Different Exposure Periods Before and After Symptom Onset JAMA Internal Medicine doi101001jamainternmed20202020
122 Resources
Cho H Ippolito D Yu YW (2020) Contact Tracing Mobile Apps for COVID-19 Privacy Considerations and Related Trade-offs httpsarxivorgpdf200311511pdf
Devakumar D Geordan S Bhopal SS Abubakar I (2020) Racism and dis-crimination in COVID-19 responses The Lancet 395(10231) 1194 doi 101016S0140-6736(20)30792-3
Drew D Nguyen L Steves C et al (2020) Rapid Implementation of Mobile Technology for Real-Time Epidemiology of COVID-19 Science published online May 5 2020
Ferretti L Wymant C Kendall M et al (2020) Quantifying SARS-CoV-2 Trans-mission Suggests Epidemic Control with Digital Contact Tracing Science doi 101126scienceabb6936
Fraser C Abeler-Doumlrner L Ferretti L et al (2020) Digital Contact Tracing Comparing the Capabilities of Centralised and Decentralised Data Archi-tectures to Effectively Suppress the COVID-19 Epidemic While Maximizing Freedom of Movement and Maintaining Privacy Preprint
Leith DJ Farrell S (2020) Coronavirus Contact Tracing Evaluating The Poten-tial Of Using Bluetooth Received Signal Strength For Proximity Detection Preprint
Jayant Limaye R Sauer M Ali J et al (2020) Building Trust While Influencing Online COVID-19 Content iIn the Social Media World The Lancet Digital Health
Mahmood S Hasan K Colder Carras M Labrique A (2020) Global Prepared-ness Aagainst COVID-19 We Must Leverage the Power of Digital Health JMIR Public Health Surveill 20206(2)e18980 DOI 10219618980
Mello M Wang CJ (2020) Ethics and Governance for Digital Disease Surveil-lance Science 11 May 2020 eabb9045 DOI 101126scienceabb9045
Park S Jeehyun Choi G Ko H (2020) Information TechnologyndashBased Tracing Strategy in Response to COVID-19 in South KoreamdashPrivacy Controversies JAMA doi101001jama20206602
Parker M Fraser C Abeler-Doumlrner L Bonsall D (2020) Ethics of Instantaneous Contract Tracing Using Mobile Phone Apps in the Control of the COVID-19 Pandemic Journal of Medical Ethics Published Online May 4 2020
Ethics and Digital Disease Detection
Aiello A Renson A Civich P (2020) Social Mediandash and Internet-Based Dis-ease Surveillance for Public Health Annual Review of Public Health 41 101ndash118 doi 101146annurev-publhealth-040119-094402
Ali J DiStefano M Coates McCall I et al (2019) Ethics of Mobile Phone Sur-
Resources 123
veys to Monitor Non-Communicable Disease Risk Factors in Low- and Middle-Income Countries A Global Stakeholder Survey Global Public Health 14(8) 1167ndash1181
Ali J Labrique A Gionfriddo K et al (2017) Ethics Considerations in Global Mobile Phone-Based Surveys of Noncommunicable Diseases A Concep-tual Exploration Journal of Medical Internet Research 19(5) e110 doi 102196jmir7326
Brownstein J Freifeld C Madoff L (2009) Digital Disease DetectionndashHar-nessing the Web for Public Health Surveillance New England Journal of Medicine 360(21) 2153ndash2157 doi 101056NEJMp0900702
Danquah LO Hasham N MacFarlane M et al (2019) Use of a Mobile Application for Ebola Contact Tracing and Monitoring in Northern Sierra Leone A Proof-of-Concept Study BMC Infectious Diseases 19 810
Degeling C Carter S van Oijen A et al (2020) Community Perspectives on the Benefits and Risks of Technologically Enhanced Communicable Disease Surveillance Systems A Report on Four Community Juries BMC Medical Ethics 21 31 doi 101186s12910-020-00474-6
DeJong B Badou G Luten J et al (2019) Ethical Considerations for Movement Mapping to Identify Disease Transmission Hotspots Emerging Infectious Diseases 25(7) e181421 doi 103201eid2507181421
Denecke K (2017) An Ethical Assessment Model for Digital Disease Detec-tion Technologies Life Sciences Society and Policy 13 16 doi 101186s40504-017-0062-x
Genevieve LD Martani A Wangmo T et al (2019) Participatory Disease Surveillance Systems Ethical Framework Journal of Medical Internet Research 21(5) e12273 doi10219612273
Gilbert G Degeling C Johnson J (2017) Communicable Disease Surveillance Ethics in the Age of Big Data and New Technology Asian Bioethics Review 11 173-187 doi 101007s41649-019-00087-1
Iwaya LH Li J Fischer-Hubner S et al (2019) E-Consent for Data Privacy Consent Management for Mobile Health Technologies in Public Health Surveys and Disease Surveillance Studies in health technology and infor-matics 264 1223-1227 doi 103233SHTI190421
Kostkova P (2018) Disease Surveillance Data Sharing for Public Health The Next Ethical Frontiers Life Sciences Society and Policy 14 16 doi 101186s40504-018-0078-x
Mahmood S Hasan K Colder Carras M Labrique A (2020) Global Prepared-ness against COVID-19 We Must Leverage the Power of Digital Health JMIR Public Health Surveillance 6(2)e18980
124 Resources
Smolinski MS Crawley AW Baltrusaitis K et al (2015) Flu Near You Crowd-sourced Symptom Reporting Spanning 2 Influenza Seasons AJPH 105(10) 2124ndash2130
Wojcik O Brownstein J Chunara R Johansson M (2014) Public Health for the People Participatory Infectious Disease Surveillance in the Digital Age Emerging Themes in Epidemiology 11 7 doi 1011861742-7622-11-7
Other Works
Bernstein J Holroyd TA Atwell JE et al (2019) Rockland Countyrsquos Proposed Ban against Unvaccinated Minors Balancing Disease Control Trust and Liberty Vaccine 37(30) 3933ndash3935
Berry SM Petzold EA Dull P et al (2017) A Response Adaptive Randomiza-tion Platform Trial for Efficient Evaluation of Ebola Virus Treatments A Model for Pandemic Response Clinical Trials 13(1) 22ndash30 doi 101177 1740774515621721
Beukenhorst AL Schultz DM McBeth J (2017) Using Smartphones for Research outside Clinical Settings How Operating Systems App Develop-ers and Users Determine Geolocation Data Quality in mHealth Studies MEDINFO 2017 Precision Healthcare through Informatics
Bourne P (2015) Confronting the Ethical Challenges of Big Data in Public Health Plos Computational Biology 11(2) e1004073 doi 101371journal pcbi1004073
Doerr M Suver C Wilbanks J (2016) Developing a Transparent Par ticipant-Navigated Electronic Informed Consent for Mobile-Mediated Research (April 22 2016) Available at SSRN httpsssrncomabstract=2769129 or httpdxdoiorg102139ssrn2769129
Dredze M Paul MJ Bergsma S Tran H (2013) Carmen A Twitter Geolocation System with Applications to Public Health Expanding the Boundaries of Health Informatics Using Artificial Intelligence Papers from the AAAI 2013 Workshop
Eckhoff PA Tatem AJ (2015) Digital methods in epidemiology can transform disease controlInternational Health Volume 7 Issue 2 March 2015 Pages 77ndash78 httpsdoiorg101093inthealthihv013
Edelstein M Lee L Herten-Crabb A Heymann D Harper D (2018) Strength-ening Global Public Health Surveillance through Data and Benefit Sharing Emerging Infectious Diseases 24(7) 1324ndash1330 doi 103201eid2407151830
Faden R Beauchamp T (1986) A History and Theory of Informed Consent Oxford University Press ISBN 9780199748655
Resources 125
Fairchild A Bayer R (2004) Ethics and the Conduct of Public Health Surveil-lance Science 303(5658) 631ndash632
Flanagan M Howe DC Nissenbaum H (2008) Embodying Values in Technol-ogy Theory and Practice In Information Technology and Moral Philoso-phy van den Hoven J amp Weckert J (eds) Cambridge Cambridge Univer-sity Press 322ndash353
Fraccaro P Beukenhorst A Sperrin M et al (2019) Digital Biomarkers from Geolocation Data in Bipolar Disorder and Schizophrenia A Systematic Review Journal of the American Medical Informatics Association Volume 26 Issue 11 November 2019 Pages 1412ndash1420
Furlanello C Merler S Menegon S et al (2002) New WEBGIS Technologies for Geo-location of Epidemiological Data An Application for the Sur-veillance of the Risk of Lyme borreliosis Disease In Giornale Italiano di Aritmologia e Cardiostimolazione Special Issue Proceedings of the 15th Int Congress ldquoThe New Frontiers of Arrhythmiasrdquo v 5 n1 Mar 2002 241ndash245
Gibson DG Wosu AC Pariyo GW et al (2019) Effect of Airtime Incentives on Response and Cooperation Rates in Non-communicable Disease Interactive Voice Response Surveys Randomised Controlled Trials in Bangladesh and Uganda BMJ Global Health 4(5)
Knobel C Bowker GC (2011) Computing Ethics Values in Design Communi-cations of the acm 54(7) 26ndash28
Lee R Cui RR Muessig KE et al (2015) Incentivizing HIVSTI Testing A Sys-tematic Review of the Literature AIDS and Behavior 18(5) 905ndash912 doi 101007s10461-013-0588-8
Lorchan LT Wyatt J (2014) mHealth and Mobile Medical Apps A Frame-work to Assess Risk and Promote Safer Use Journal of Medical Internet Research 16(9) e210 doi 102196jmir3133
Mathews S McShea M Hanley C et al (2019) Digital Health A Path to Vali-dation Digital Medicine 238 doi 101038s41746-019-0111-3
Moore S Tasse A Thorogood A et al (2017) Consent Processes for Mobile App Mediated Research Systematic Review JMIR mHealth amp uHealth 5(8) e126 DOI 102196mhealth7014
Pallman P Bedding AW Choodari-Oskooei B et al (2018) Adaptive Designs in Clinical Trials Why Use Them and How to Run and Report Them BMC Medicine 16(29)
Rennie S Buchbinder M Juengst E et al (2020) Scraping the Web for Public Health Gains Ethical Considerations from a lsquoBig Datarsquo Research Project on HIV and Incarceration Public Health Ethics doi 101093phephaa006
126 Resources
Rithalia A McDaid C Suekarran S (2009) Impact of Presumed Consent for Organ Donation on Donation Rates A Systematic Review BMJ 2009 338
Singer E Ye C (2012) The Use and Effects of Incentives in Surveys The ANNALS of the American Academy of Political and Social Science 645(1) 112ndash141
Vayena E Blassime A (2018) Health Research with Big Data Time for Systemic Oversight J Law Med Ethics 2018 Mar 46(1) 119ndash129
Vayena E Mastroianni A Kahn J (2012) Ethical Issues in Health Research with Novel Online Sources American Journal of Public Health 102(12) 2225ndash2230 doi 102105AJPH2012300813
Vayena E Salatheacute M Madoff L Brownstein J (2015) Ethical Challenges of Big Data in Public Health Plos Computational Biology 11(2) e1003904 doi 101371journalpcbi1003904
Xafis V Schaefer GO Labude MK et al (2019) An Ethics Framework for Big Data in Health and Research Asian Bioethics Review volume 11 pages 227ndash254(2019)
Works Cited
127
Africa CDC 2020 ldquoGuidance on Contact Tracing for COVID-19 Pandemicrdquo Manuals Guidelines amp Frameworks African Union Africa CDC httpsafricacdcorgdownloadguidance-on-contact-tracing-for-covid-19- pandemic
Aiello Allison E Audrey Renson and Paul N Zivich 2020 ldquoSocial Mediandash and Internet-Based Disease Surveillance for Public Healthrdquo Annual Review of Public Health 41 (1) 101ndash18 httpsdoiorg101146annurev-publhealth-040119-094402
Ali Joseph Michael J DiStefano Iris Coates McCall Dustin G Gibson Gulam Muhammed Al Kibria George W Pariyo Alain B Labrique and Adnan A Hyder 2019 ldquoEthics of Mobile Phone Surveys to Monitor Non-Communicable Disease Risk Factors in Low- and Middle-Income Countries A Global Stakeholder Surveyrdquo Global Public Health 14 (8) 1167ndash81 httpsdoiorg1010801744169220191566482
Ali Joseph Alain B Labrique Kara Gionfriddo George Pariyo Dustin G Gibson Bridget Pratt Molly Deutsch-Feldman and Adnan A Hyder 2017 ldquoEthics Considerations in Global Mobile Phone-Based Surveys of Noncommunicable Diseases A Conceptual Explorationrdquo Journal of Medi-cal Internet Research 19 (5) e110 httpsdoiorg102196jmir7326
Allen Danielle Julius Krein Ganesh Sitaraman and E Glen Weyl 2020 ldquoNational Covid-19 Testing Action Planrdquo The Rockefeller Foundation httpswwwrockefellerfoundationorgwp-contentuploads202004TheRockefellerFoundation_WhitePaper_Covid19_4_22_2020pdf
Altmann Samuel Luke Milsom Hannah Zillessen Raffaele Blasone Frederic Gerdon Ruben Bach Frauke Kreuter Daniele Nosenzo Severine Tous-saert and Johannes Abeler 2020 ldquoAcceptability of App-Based Contact Tracing for COVID-19 Cross-Country Survey Evidencerdquo MedRxiv May 2020050520091587 httpsdoiorg1011012020050520091587
Anderson Monica and Brooke Auxier 2020 ldquoMost Americans Donrsquot Think Cellphone Tracking Will Help Limit COVID-19 Are Divided on Whether
128 Works Cited
Itrsquos Acceptablerdquo Pew Research Center (blog) April 16 2020 https wwwpewresearchorgfact-tank20200416most-americans-dont-think-cellphone-tracking-will-help-limit-covid-19-are-divided-on-whether-its- acceptable
Anderson Monica and Andrew Perrin 2017 ldquoDisabled Americans Less Likely to Use Technologyrdquo Pew Research Center (blog) April 7 2017 httpswwwpewresearchorgfact-tank20170407disabled-americans- are-less-likely-to-use-technology
Apple and Google nd ldquoPrivacy-Preserving Contact Tracingrdquo Apple Accessed May 19 2020 httpswwwapplecomcovid19contacttracing
ASTHO 2012 ldquoPublic Health Collection Use Sharing and Protection of Informationrdquo Issue Brief ASTHO Legal Preparedness Series Association of State and Territorial Health Officials httpswwwasthoorgProgramsPreparednessPublic-Health-Emergency-LawPublic-Health-and- Information-Sharing-ToolkitCollection-Use-Sharing-and-Protection- Issue-Brief
Auxier Brooke Lee Rainie Monica Anderson Andrew Perrin Madhu Kumar and Erica Turner 2019 ldquoAmericans and Privacy Concerned Confused and Feeling Lack of Control Over Their Personal Informationrdquo Pew Research Center Internet Science amp Tech (blog) November 15 2019 httpswwwpewresearchorginternet20191115americans-and-privacy- concerned-confused-and-feeling-lack-of-control-over-their-personal-infor mation
Barth Susanne and Menno D T de Jong 2017 ldquoThe Privacy Paradox ndash Inves-tigating Discrepancies between Expressed Privacy Concerns and Actual Online Behavior ndash A Systematic Literature Reviewrdquo Telematics and Infor-matics 34 (7) 1038ndash58 httpsdoiorg101016jtele201704013
Bernstein Justin Taylor A Holroyd Jessica E Atwell Joseph Ali and Rupali J Limaye 2019 ldquoRockland Countyrsquos Proposed Ban against Unvaccinated Minors Balancing Disease Control Trust and Libertyrdquo Vaccine June httpsdoiorg101016jvaccine201905093
Berry Scott M Elizabeth A Petzold Peter Dull Nathan M Thielman Coleen K Cunningham G Ralph Corey Micah T McClain et al 2016 ldquoA Response Adaptive Randomization Platform Trial for Efficient Evalua-tion of Ebola Virus Treatments A Model for Pandemic Responserdquo Clinical Trials (London England) 13 (1) 22ndash30 httpsdoiorg101177 1740774515621721
Beukenhorst AL DM Schultz J McBeth R Lakshminarayana JC Sergeant and WG Dixon 2017 ldquoUsing Smartphones for Research Outside Clinical
Works Cited 129
Settings How Operating Systems App Evelopers and Users Determine Geolocation Data Quality in MHealth Studiesrdquo In MEDINFO 2017 Precision Healthcare through Infomatics 10ndash14 IOS Press httpebooks iospressnlvolumearticle48095
Bradshaw Tim 2020 ldquo2 Billion Phones Cannot Use Google and Apple Con-tact-Tracing Tech | Ars Technicardquo Ars Technica April 20 2020 httpsarstechnicacomtech-policy2020042-billion-phones-cannot-use-google-and-apple-contract-tracing-tech
Canca Cansu 2020 ldquoWhy lsquoMandatory Privacy-Preserving Digital Contact Tracingrsquo Is the Ethical Measure against COVID-19rdquo Medium May 4 2020 httpsmediumcomcansucancawhy-mandatory-privacy- preserving-digital-contact-tracing-is-the-ethical-measure-against- covid-19-a0d143b7c3b6
Cavalier Robert J ed 2011 Approaching Deliberative Democracy Theory and Practice Carnegie Mellon University Press
Cavoukian Ann 2010 ldquoPrivacy by Design The 7 Foundational Principlesrdquo Ontario Canada Information and Privacy Commissioner of Canada httpsiapporgmediapdfresource_centerpbd_implement_7found_ principlespdf
CDC 2005 ldquoVI Plan for Surveillance of Contacts of SARS Case Supplement B SARS Surveillancerdquo Centers for Disease Control and Prevention httpswwwcdcgovsarsguidanceb-surveillancecontactshtml
mdashmdashmdash 2020a ldquoContact Tracingrdquo Get and Keep America Open Supporting States Tribes Localities and Territories Centers for Disease Control and Prevention httpswwwcdcgovcoronavirus2019-ncovphpopen-americacontact-tracinghtml
mdashmdashmdash 2020b ldquoContact Tracing Part of a Multipronged Approach to Fight the COVID-19 Pandemicrdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention httpswwwcdcgovcoronavirus2019- ncovphpprinciples-contact-tracinghtml
mdashmdashmdash 2020c ldquoInterim Guidance for Businesses and Employersrdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention httpswwwcdcgovcoronavirus2019-ncovcommunityguidance-business- responsehtml
mdashmdashmdash 2020d ldquoOperational Considerations for the Identification of Healthcare Workers and Inpatients with Suspected COVID-19 in Non-US Healthcare Settingsrdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention httpswwwcdcgovcoronavirus2019-ncovhcpnon-us-settingsguidance-identify-hcw-patientshtml
130 Works Cited
mdashmdashmdash 2020e ldquoPreliminary Criteria for the Evaluation of Digital Contact Trac-ing Tools for COVID-19rdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention httpswwwcdcgovcoronavirus 2019-ncovdownloadsphpprelim-eval-criteria-digital-contact-tracingpdf
mdashmdashmdash 2020f ldquoLegal Authorities for Isolation and Quarantinerdquo Centers for Disease Control and Prevention httpswwwcdcgovquarantine aboutlawsregulationsquarantineisolationhtml
mdashmdashmdash 2020g ldquoDiscontinuation of Isolation for Persons with COVID-19 Not in Healthcare Settingsrdquo Interim Guidance Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention httpswww cdcgovcoronavirus2019-ncovhcpdisposition-in-home-patientshtml
mdashmdashmdash 2020h ldquoCOVID-19 Provisional CountsndashWeekly Updates by Select Demographic and Geographic Characteristicsrdquo CDC National Vital Statis-tics System httpswwwcdcgovnchsnvssvsrrcovid_weeklyindexhtm
Center for Survey Measurement (CSM) MEMORANDUM FOR Associate Directorate for Research and Methodology (ADRM) 2017 ldquoRespondent Confidentiality Concernsrdquo September 20 2017 httpswww2censusgovcacnacmeetings2017-11Memo-Regarding-Respondent-Confidentiality- Concernspdf
Cheng Hao-Yuan Shu-Wan Jian Ding-Ping Liu Ta-Chou Ng Wan-Ting Huang and Hsien-Ho Lin 2020 ldquoContact Tracing Assessment of COVID-19 Transmission Dynamics in Taiwan and Risk at Different Exposure Periods Before and After Symptom Onsetrdquo JAMA Internal Medicine May httpsdoiorg101001jamainternmed20202020
Cochrane Emily Claire Cain Miller and Jim Tankersley 2020 ldquoTrump Admin-istration Scales Back Paid Leave in Coronavirus Relief Lawrdquo The New York Times April 2 2020 sec US httpswwwnytimescom20200402uspoliticscoronavirus-paid-leavehtml
Crocker Andrew Kurt Opsahl and Bennett Cyphers 2020 ldquoThe Challenge of Proximity Apps For COVID-19 Contact Tracingrdquo Electronic Fron-tier Foundation April 10 2020 httpswwwefforgdeeplinks202004challenge-proximity-apps-covid-19-contact-tracing
Danquah Lisa O Nadia Hasham Matthew MacFarlane Fatu E Conteh Fatoma Momoh Andrew A Tedesco Amara Jambai David A Ross and Helen A Weiss 2019 ldquoUse of a Mobile Application for Ebola Contact Tracing and Monitoring in Northern Sierra Leone A Proof-of-Concept Studyrdquo BMC Infectious Diseases 19 (1) 810 httpsdoiorg101186s12879-019-4354-z
de Jong Bouke C Badou M Gaye Jeroen Luyten Bart van Buitenen Emman-
Works Cited 131
uel Andreacute Conor J Meehan Cian OrsquoSiochain et al 2019 ldquoEthical Considerations for Movement Mapping to Identify Disease Transmission Hotspotsrdquo Emerging Infectious Diseases 25 (7) httpsdoiorg103201eid2507181421
Devakumar Delan Geordan Shannon Sunil S Bhopal and Ibrahim Abu-bakar 2020 ldquoRacism and Discrimination in COVID-19 Responsesrdquo Lancet (London England) 395 (10231) 1194 httpsdoiorg101016S0140-6736(20)30792-3
Dixit Pranav 2020 ldquoIndiarsquos Contact Tracing App Is All But Mandatory So This Programmer Hacked It So That He Always Appears Saferdquo BuzzFeed News May 12 2020 httpswwwbuzzfeednewscomarticlepranavdixitindia-aarogya-setu-hacked
Doerr Megan Christine Suver and John Wilbanks 2016 ldquoDeveloping a Trans-parent Participant-Navigated Electronic Informed Consent for Mobile- Mediated Researchrdquo SSRN Scholarly Paper ID 2769129 Rochester NY Social Science Research Network httpsdoiorg102139ssrn2769129
Dredze Mark Michael J Paul Shane Bergsma and Hieu Tran 2013 ldquoCarmen A Twitter Geolocation System with Applications to Public Healthrdquo AAAI Workshops Workshops at the Twenty-Seventh AAAI Conference on Artifi-cial Intelligence httpswwwaaaiorgocsindexphpWSAAAIW13paperview70856497
Eckhoff Philip A and Andrew J Tatem 2015 ldquoDigital Methods in Epidemi-ology Can Transform Disease Controlrdquo International Health 7 (2) 77ndash78 httpsdoiorg101093inthealthihv013
EEOC 2020 ldquoWhat You Should Know About COVID-19 and the ADA the Rehabilitation Act and Other EEO Lawsrdquo US Equal Employment Opportunity Commission May 7 2020 httpswwweeocgovwyskwhat-you-should-know-about-covid-19-and-ada-rehabilitation-act-and-other-eeo-laws
Electronic Privacy Information Center Testimony to Congress 2020 ldquoEPIC to Congress Establish Privacy Safeguards for Digital Contact Tracingrdquo April 15 2020 httpsepicorgtestimonycongressEPIC-HEC-Contact-Tracing-Apr2020pdf
Elliott Douglas J Ana Kreacic Lorenzo Milans del Bosch and Lisa Quest 2020 ldquoData-Sharing in the Time of Coronavirusrdquo Oliver Wyman Forum Accessed May 19 2020 httpswwwoliverwymanforumcomfuture-of-data2020aprdata-sharing-in-the-time-of-coronavirushtml
Faden Ruth R and Tom L Beauchamp 1986 A History and Theory of Informed Consent Oxford University Press
132 Works Cited
FCC and FTC 2017 ldquoFCC-FTC Consumer Protection Memorandum of Understandingrdquo Federal Communications Commission and Federal Trade Commission httpswwwftcgovsystemfilesdocumentscooperation_agreements151116ftcfcc-moupdf
Ferretti Luca Chris Wymant Michelle Kendall Lele Zhao Anel Nurtay Lucie Abeler-Doumlrner Michael Parker David Bonsall and Christophe Fraser 2020 ldquoQuantifying SARS-CoV-2 Transmission Suggests Epidemic Control with Digital Contact Tracingrdquo Science 368 (6491) httpsdoiorg101126scienceabb6936
Fishkin James S and Peter Laslett eds 2003 Debating Deliberative Democ-racy Blackwell Publisher Ltd httpsonlinelibrarywileycomdoibook 1010029780470690734
Flanagan Mary Daniel C Howe and Helen Nissenbaum 2008 ldquoEmbodying Values in Technology Theory and Practicerdquo In Information Technology and Moral Philosophy edited by Jeroen van den Hoven and John Weckert 322ndash53 Cambridge University Press
Fraccaro Paolo Anna Beukenhorst Matthew Sperrin Simon Harper Jasper Palmier-Claus Shocircn Lewis Sabine N Van der Veer and Niels Peek 2019 ldquoDigital Biomarkers from Geolocation Data in Bipolar Disorder and Schizophrenia A Systematic Reviewrdquo Journal of the American Medical Informatics Association 26 (11) 1412ndash20 httpsdoiorg101093jamiaocz043
Fraser Christophe Lucie Abeler-Doumlrner Luca Ferretti Michael Parker Michelle Kendall and David Bonsall 2020 ldquoDigital Contact Tracing Comparing the Capabilities of Centralised and Decentralised Data Architectures to Effectively Suppress the COVID-19 Epidemic Whilst Maximising Freedom of Movement and Maintaining Privacyrdquo httpsgithubcomBDI-pathogenscovid-19_instant_tracingblobmasterCentralised20and20decen-tralised20systems20for20contact20tracingpdf
FTC 2012 ldquoProtecting Consumer Privacy in an Era of Rapid Changerdquo FTC Report Recommendations for Businesses and Policymakers Federal Trade Commission httpswwwftcgovsitesdefaultfilesdocumentsreports federal-trade-commission-report-protecting-consumer-privacy- era-rapid-change-recommendations120326privacyreportpdf
mdashmdashmdash 2020 ldquoPrivacy amp Data Security Update 2019rdquo Federal Trade Commis-sion httpswwwftcgovsystemfilesdocumentsreportsprivacy-data- security-update-20192019-privacy-data-security-report-508pdf
Furlanello Cesare Stefano Merler Stefano Menegon Sebastiano Mancuso and Gianni Bertiato 2002 ldquoNew WEBGIS Technologies for Geo-Location of
Works Cited 133
Epidemiological Data An Application for the Surveillance of the Risk of Lyme Borreliosis Diseaserdquo Giornale Italiano Di Aritmologia e Cardiosti-molazione Special Issue Proceedings of the 15th Int Congress ldquoThe New Frontiers of Arrhythmiasrdquo 5 (1) 241ndash45
Gan Nectar and David Culver 2020 ldquoChina Is Fighting the Coronavirus with a Digital QR Code Herersquos How It Worksrdquo CNN Business April 16 2020 httpswwwcnncom20200415asiachina-coronavirus-qr-code-intl-hnkindexhtml
Gibson Dustin G Adaeze C Wosu George William Pariyo Saifuddin Ahmed Joseph Ali Alain B Labrique Iqbal Ansary Khan Elizeus Rutebemberwa Meerjady Sabrina Flora and Adnan A Hyder 2019 ldquoEffect of Airtime Incentives on Response and Cooperation Rates in Non-Communicable Disease Interactive Voice Response Surveys Randomised Controlled Trials in Bangladesh and Ugandardquo BMJ Global Health 4 (5) e001604 httpsdoiorg101136bmjgh-2019-001604
Giglio Mike 2020 ldquoWould You Sacrifice Your Privacy to Get Out of Quarantinerdquo The Atlantic April 22 2020 httpswwwtheatlanticcompoliticsarchive202004coronavirus-pandemic-privacy-civil-liberties-911 609172
Guariglia Matthew 2020 ldquoThe Dangers of COVID-19 Surveillance Proposals to the Future of Protestrdquo Electronic Frontier Foundation April 29 2020 httpswwwefforgdeeplinks202004some-covid-19-surveillance-propos als-could-harm-free-speech-after-covid-19
Hadavas Chloe 2020 ldquoHow Effective Are Contact Tracing Appsrdquo Slate Maga-zine May 13 2020 httpsslatecomtechnology202005contact- tracing-apps-less-effective-icelandhtml
Hamilton Isobel Asher 2020 ldquoPoland Made an App That Forces Coronavirus Patients to Take Regular Selfies to Prove Theyrsquore Indoors or Face a Police Visitrdquo Business Insider March 23 2020 httpswwwbusinessinsidercompoland-app-coronavirus-patients-mandaotory-selfie-2020-3
Hargittai Eszter Minh Hao Nguyen Jaelle Fuchs Jonathan Gruber Will Marler Amanda Hunsaker and Goumlkccedile Karaoglu 2020 ldquoCovid-19 Study on Digital Media and the Coronavirus Pandemicrdquo Internet Use and Society Division Institute of Communication and Media Research University of Zurich httpwebuseorgcovid
Hargittai Eszter and Elissa Redmiles 2020 ldquoWill Americans Be Willing to Install COVID-19 Tracking Appsrdquo Scientific American Blog Net-work April 28 2020 httpsblogsscientificamericancomobservationswill-americans-be-willing-to-install-covid-19-tracking-apps
134 Works Cited
Hart Vi Divya Siddarth Bethan Cantrell Lila Tretikov Peter Eckersley John Langford Scott Leibrand et al 2020 ldquoOutpacing the Virus Dig-ital Response to Containing the Spread of COVID-19 While Mitigat-ing Privacy Risksrdquo Whitepaper 5 COVID-19 Rapid Response Impact Initiative Edmond J Safra Center for Ethics httpsdrivegooglecomfiled1vIN2AX-DDNW-S0aHq8xs0RJ2jkR_CckXview
Hemming K T P Haines P J Chilton A J Girling and R J Lilford 2015 ldquoThe Stepped Wedge Cluster Randomised Trial Rationale Design Anal-ysis and Reportingrdquo BMJ 350 (February) httpsdoiorg101136bmjh391
Hendrix Steve and Ruth Eglash 2020 ldquoIsrael Is Using Cellphone Surveil-lance to Warn Citizens You May Already Be Infectedrdquo Washington Post Accessed May 19 2020 httpswwwwashingtonpostcomworldmiddle_eastisrael-is-using-cellphone-surveillance-to-warn-citizens-you-may- already-be-infected2020031968267294-69e7-11ea-b199-3a9799c54 512_storyhtml
Heneghan Carl Jon Brassey and Tom Jefferson 2020 ldquoCOVID-19 What Proportion Are Asymptomaticrdquo Centre for Evidence-Based Medi-cine httpswwwcebmnetcovid-19covid-19-what-proportion-are- asymptomatic
HHS 2019 ldquoPublic Health Emergency Declarationrdquo Public Health Emergency Accessed May 19 2020 httpswwwphegovPreparednesslegalPagesphedeclarationaspx
Hickey Matt 2014 ldquoCarriers Can Now Install Apps On Android Handsets Without Customersrsquo Permissionrdquo Forbes December 1 2014 httpswww forbescomsitesmatthickey20141201carriers-can-now-install- apps-on-android-handsets-without-customers-permission
Hinch Robert Will Probert Anel Nurtay Michelle Kendall Chris Wymant Matthew Hall Katrina Lythgoe et al 2020 ldquoEffective Configurations of a Digital Contact Tracing App A Report to NHSXrdquo fileUsersameliahoodDownloadsReport20-20Effective20App20Configurations20(1)pdf
Ingram David 2020 ldquoApple Google Push Makers of Coronavirus Apps Not to Record User Locationrdquo May 4 2020 NBC News Accessed May 19 2020 httpswwwnbcnewscomtechtech-newscoronavirus-apps-won-t-be-able- record-users-location-apple-n1199586
Iwaya Leonardo H Jane Li Simone Fischer-Huumlbner Rose-Mharie Aringhlfeldt and Leonardo A Martucci 2019 ldquoE-Consent for Data Privacy Consent Management for Mobile Health Technologies in Public Health Surveys and
Works Cited 135
Disease Surveillancerdquo Studies in Health Technology and Informatics 264 (August) 1223ndash27 httpsdoiorg103233SHTI190421
ldquoJoint Statement on Contact Tracingrdquo 2020 April 19 httpscryptobriefingcomwp-contentuploads202004Joint-Statement-from-Researcherspdf
Kahn Gilmor Daniel 2020 ldquoPrinciples for Technology-Assisted Contact- Tracingrdquo White Paper American Civil Liberties Union httpswwwaclu orgreportaclu-white-paper-principles-technology-assisted-contact-tracing
Kim Max S 2020 ldquoSeoulrsquos Radical Experiment in Digital Contact Tracingrdquo The New Yorker April 17 2020 httpswwwnewyorkercomnewsnews-deskseouls-radical-experiment-in-digital-contact-tracing
Kim Nemo 2020 ldquolsquoMore Scary than Coronavirusrsquo South Korearsquos Health Alerts Expose Private Livesrdquo The Guardian March 6 2020 sec World news httpswwwtheguardiancomworld2020mar06more-scary- than-coronavirus-south-koreas-health-alerts-expose-private-lives
Kirzinger Ashley Liz Hamel Cailey Muntildeana Audrey Kearney and Mollyann Brodie 2020 ldquoKFF Health Tracking Poll ndash Late April 2020 Coronavirus Social Distancing and Contact Tracingrdquo Kaiser Family Foundation (blog) April 24 2020 httpswwwkfforgcoronavirus-covid-19issue-briefkff-health-tracking-poll-late-april-2020
Knobel Cory and Geoffrey C Bowker 2011 ldquoComputing Ethics Value in Designrdquo Communications of the ACM 54 (7) 26ndash28 httpsdoiorgdoi10114519657241965735
Lee Ramon Rosa R Cui Kathryn E Muessig Harsha Thirumurthy and Joseph D Tucker 2014 ldquoIncentivizing HIVSTI Testing A Systematic Review of the Literaturerdquo AIDS and Behavior 18 (5) 905ndash12 https doiorg101007s10461-013-0588-8
Leprince-Ringuet Daphne 2020 ldquoContact-Tracing Apps Why the NHS Said No to Apple and Googlersquos Planrdquo ZDNet April 28 2020 httpswwwzdnetcomarticlecontact-tracing-apps-why-the-nhs-said-no-to-apple-and- googles-plan
Lovejoy Ben 2020 ldquoNHS Contact Tracing App Hands-on as the Test Goes Liverdquo 9to5Mac (blog) May 7 2020 https9to5maccom20200507nhs-contact-tracing
Mahmood Sultan Khaled Hasan Michelle Colder Carras and Alain Labri-que 2020 ldquoGlobal Preparedness Against COVID-19 We Must Leverage the Power of Digital Healthrdquo JMIR Public Health and Surveillance 6 (2) e18980 httpsdoiorg10219618980
Mathews Simon C Michael J McShea Casey L Hanley Alan Ravitz Alain B Labrique and Adam B Cohen 2019 ldquoDigital Health A Path to
136 Works Cited
Validationrdquo Npj Digital Medicine 2 (1) 1ndash9 httpsdoiorg101038s41746-019-0111-3
Mello By Michelle M and C Jason Wang 2020 ldquoEthics and Governance for Digital Disease Surveillancerdquo Science May httpsdoiorg101126science abb9045
Mills Rodrigo Chris 2020 ldquoDigital Contact Tracing Is Becoming Available but Is It Effectiverdquo The Hill May 7 2020 httpsthehillcompolicytechnolo-gy496498-digital-contact-tracing-is-becoming-available-but-is-it-effective
MIT nd ldquoPrivate Kit Safe Paths Privacy-by-Design Covid19 Solutions Using GPS+Bluetooth for Citizens and Public Health Officialsrdquo Safepaths Accessed May 20 2020 httpssafepathsmitedu
ldquoMobile Cellular Subscriptions (per 100 People)rdquo 2018 World Telecommuni-cationICT Development Report International Telecommunication Union httpsdataworldbankorgindicatorITCELSETSP2
Moore Sarah Anne-Marie Tasseacute Adrian Thorogood Ingrid Winship Marsquon Zawati and Megan Doerr 2017 ldquoConsent Processes for Mobile App Mediated Research Systematic Reviewrdquo JMIR MHealth and UHealth 5 (8) e126 httpsdoiorg102196mhealth7014
Morse Jack 2020 ldquoNorth Dakota Launched a Contact-Tracing App Itrsquos Not Going Wellrdquo Mashable Accessed May 19 2020 httpsmashablecomarticlenorth-dakota-contact-tracing-app
Muller Robert T 2020 ldquoCOVID-19 Brings a Pandemic of Conspiracy Theo-riesrdquo Psychology Today April 24 2020 httpswwwpsychologytodaycomblogtalking-about-trauma202004covid-19-brings-pandemic-conspiracy- theories
Mulligan Stephen P and Chris D Linebaugh 2019 ldquoData Protection Law An Overviewrdquo R45631 Congressional Research Service httpscrsreports congressgovproductpdfRR45631
National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research 1979 ldquoThe Belmont Report Ethical Principles and Guidelines for the Protection of Human Subjects of Researchrdquo US Dept of Health amp Human Services Office of Human Research Protections httpswwwhhsgovohrpregulations-and-policybelmont-reportread-the-belmont-reportindexhtml
NDDoH 2020 ldquoNorth Dakota Launches Care19 App to Combat COVID-19rdquo North Dakota Department of Health April 7 2020 httpswwwhealth ndgovnewsnorth-dakota-launches-care19-app-combat-covid-19
NYC DOH 2020 ldquoAge-Adjusted Rates of Lab Confirmed COVID-19 Non-hospitalized Cases Estimated Non-Fatal Hospitalized Cases and Persons Known to Have Died per 100000 by RaceEthnicity Grouprdquo New York
Works Cited 137
City Department of Health httpswww1nycgovassetsdohdownloadspdfimmcovid-19-deaths-race-ethnicity-04242020-1pdf
OrsquoRielly Michael 2018 ldquoFCC Regulatory Free Arenardquo Federal Communica-tions Commission June 1 2018 httpswwwfccgovnews-eventsblog 20180601fcc-regulatory-free-arena
OSHA 2020 ldquoGuidance on Preparing Workplaces for COVID-19rdquo OSHA 3990-03 2020 Department of Labor Occupational Safety and Health Act of 1970 httpswwwoshagovPublicationsOSHA3990pdf
Pallmann Philip Alun W Bedding Babak Choodari-Oskooei Munyaradzi Dimairo Laura Flight Lisa V Hampson Jane Holmes et al 2018 ldquoAdaptive Designs in Clinical Trials Why Use Them and How to Run and Report Themrdquo BMC Medicine 16 (1) 29 httpsdoiorg101186s12916-018-1017-7
Pew Research Center 2017 ldquoUS Muslims Concerned About Their Place in Society but Continue to Believe in the American Dreamrdquo Religion amp Public Life Pew Research Center httpswwwpewforumorg20170726findings-from-pew-research-centers-2017-survey-of-us-muslims
mdashmdashmdash 2020 ldquoDemographics of Mobile Device Ownership and Adoption in the United Statesrdquo Internet amp Technology Pew Research Center Accessed May 19 2020 httpswwwpewresearchorginternetfact-sheetmobile
PHLS 2002 ldquoPrinciples of the Ethical Practice of Public Healthrdquo Public Health Leadership Society httpswwwaphaorg-mediafilespdfmembergroupsethicsethics_brochureashx
PIH 2020a ldquoPart I Testing Contact Tracing and Community Management of COVID-19rdquo PIH Guide | COVID-19 STOP COVID Partners in Health httpswwwpihorgsitesdefaultfiles2020-04PIH_Guide_COVID_Part_I_Testing_Tracing_Community_Managment_4_4pdf
mdashmdashmdash 2020b ldquoBox It Inrdquo PIH Guide | COVID-19 Partners in Health httpspreventepidemicsorgcovid19resourcesbox-it-in
Resolve to Save Lives 2020 ldquoCOVID-19 Contact Tracing Playbookrdquo Vital Strategies httpscontacttracingplaybookresolvetosavelivesorg
Reston Maeve Kristina Sgueglia and Cheri Mossburg 2020 ldquoGovernors on East and West Coasts Form Pacts to Decide When to Reopen Economiesrdquo CNN Politics April 13 2020 httpswwwcnncom20200413politicsstates-band-together-reopening-plansindexhtml
Rithalia Amber Catriona McDaid Sara Suekarran Lindsey Myers and Amanda Sowden 2009 ldquoImpact of Presumed Consent for Organ Donation on Donation Rates A Systematic Reviewrdquo BMJ 338 (January) httpsdoiorg101136bmja3162
Rodrigues Rafaela Alina Husain Amanda Couture-Carron Leslye E Orloff
138 Works Cited
and Nawal H Ammar 2018 ldquoPromoting Access to Justice for Immigrant and Limited English Proficient Crime Victims in an Age of Increased Immi-gration Enforcement Initial Report from a 2017 National Surveyrdquo Wash-ington DC National Immigrant Womenrsquos Advocacy Project American University Washington College of Law httpniwaplibrarywclamericaneduwp-contentuploadsImmigrant-Access-to-Justice-National-Reportpdf
Simpson Erin and Adam Conner 2020 ldquoDigital Contact Tracing to Contain the Coronavirusrdquo Center for American Progress httpswww americanprogressorgissuestechnology-policynews20200422483521digital-contact-tracing-contain-coronavirus
Singer Eleanor and Cong Ye 2013 ldquoThe Use and Effects of Incentives in Surveysrdquo The ANNALS of the American Academy of Political and Social Science 645 (1) 112ndash41 httpsdoiorg1011770002716212458082
Thornton Rebecca L 2008 ldquoThe Demand for and Impact of Learning HIV Statusrdquo The American Economic Review 98 (5) 1829ndash63 httpsdoi org101257aer9851829
US DOE 2009 ldquoState Regulation of Private Schoolsrdquo US Dept of Education Office of Non-Public Education httpswww2edgovadminscommchoiceregprivschlregprivschlpdf
Valentino-DeVries Jennifer 2020 ldquoCellphone Carriers Face $200 Million Fine for Not Protecting Location Datardquo The New York Times February 28 2020 sec Technology httpswwwnytimescom20200228technologyfcc-cellphones-location-data-fineshtml
Valentino-DeVries Jennifer Natasha Singer and Aaron Krolik 2020 ldquoA Scramble for Virus Apps That Do No Harmrdquo The New York Times April 29 2020 sec Business httpswwwnytimescom20200429businesscoronavirus-cellphone-apps-contact-tracinghtml
Washington PostndashUMD ldquoWashington Post-University of Maryland National Pollrdquo 2020 Washington Post April 21 2020 httpswwwwashingtonpostcomcontextwashington-post-university-of-maryland-national-poll-april-21-26-20203583b4e9-66be-4ed6-a457-f6630a550ddf
Watson Crystal Anita Cicero James Blumenstock and Michael Fraser 2020 ldquoA National Plan to Enable Comprehensive COVID-19 Case Finding and Contact Tracing in the USrdquo Johns Hopkins Bloomberg School of Public Health Center for Health Security and the Association of State and Terri-torial Health Officials httpswwwcenterforhealthsecurityorgour-workpubs_archivepubs-pdfs2020200410-national-plan-to-contact-tracingpdf
Works Cited 139
WHO 2017 ldquoWHO Guidelines on Ethical Issues in Public Health Surveillancerdquo World Health Organization httpswwwwhointethicspublicationspublic-health-surveillanceen
mdashmdashmdash 2020 ldquoCoronavirus Disease 2019 (COVID-19)rdquo Situation Report 73 World Health Organization httpswwwwhointdocsdefault-sourcecoro-navirusesituation-reports20200402-sitrep-73-covid-19pdf
- _iy90yrtqgtxz
- _GoBack
- Acknowledgments
- Preface
- Lead Authors and Contributors
- Acronyms and Abbreviations
- Summary
-
- Introduction
- DCTT Features Functions and Potential Applications
- Summary of Recommendations
- Summary of Analysis
-
- Introduction
-
- Guiding Principles for the Use of Digital Public Health Technologies for Pandemic Response
-
- Public Health Perspective
-
- Types of Information Collected through Contact Tracing
- How Contact Tracing Information Informs Public Health Action
- Characteristics That Make Data Useful to Public Health for Reducing Disease Transmission
-
- Digital Technology and Contact Tracing
-
- Characteristics of SARS-CoV-2 Relevant to Candidate Digital Solutions
- Previously Existing Contact Tracing Technologies
- Introduction of Novel Digital Contact Tracing Technologies
- Relevant Differences between Manual and Digital Contact Tracing
-
- Ethics of Designing and Using DCTT
-
- Justifying the Use of DCTT Systems
- Monitoring and Evaluating Technologies to Inform Policy and Practice
- Public Trust and Public Attitudes
- Designing Flexible Technology to Maximize Public Health Utility While Respecting Other Values
- Policy Positions to Advance Widespread Use of Digital Contact Tracing Technologies
- Disclosure and AuthorizationConsent
- Promoting Equity and Fairness
- Instituting Transparent Governance and Oversight
-
- Legal Considerations
-
- Data Privacy and Data Security Laws
- Health Information Privacy
- Labor and Employment Privacy Rights
- Constitutional Privacy Rights
- Consent
- Anti-discrimination and Individual Freedom Laws
-
- Recommendations
-
- Public Health
- Ethics
- Legislative
-
- Resources
-
- US Government Response
- Other Governmental and Nongovernmental Organizations
- Digital Contact Tracing Experiences from Other Countries
- Specific Digital ProductsApps
- Polling
- Popular Press
- Commentaries
- Academic Literature
-
- Works Cited
-
DIGITAL CONTACT TRACING FOR PANDEMIC RESPONSE
DIGITAL CONTACT TRACING FOR PANDEMIC RESPONSEEthics and Governance Guidance
Johns Hopkins University PressBaltimore
Edited by Jeffrey P Kahn PhD MPH Chair
Johns Hopkins Project on Ethics and Governance of Digital Contact Tracing Technologies
copy 2020 Johns Hopkins University Press
This work is also available in an Open Access edition which is licensed
under a Creative Commons AttributionndashNonCommercialndashNoDerivatives 40
International License httpscreativecommonsorglicensesby-nc-nd40
All rights reserved Published 2020
Printed in the United States of America on acid-free paper
9 8 7 6 5 4 3 2 1
Johns Hopkins University Press
2715 North Charles Street
Baltimore Maryland 21218-4363
wwwpressjhuedu
Library of Congress Cataloging-in-Publication Data is available
ISBN 978-1-4214-4061-3 (paperback acid-free paper)
ISBN 978-1-4214-4062-0 (ebook)
ISBN 978-1-4214-4063-7 (ebook open access)
Special discounts are available for bulk purchases of this book For more information
please contact Special Sales at specialsalespressjhuedu
Johns Hopkins University Press uses environmentally friendly book materials
including recycled text paper that is composed of at least 30 percent post-
consumer waste whenever possible
Contents
Lead Authors and Contributors vii
Preface ix
Acknowledgments xi
Acronyms and Abbreviations xiii
Summary 1
Introduction 1
DCTT Features Functions and Potential Applications 3
Summary of Recommendations 9
Summary of Analysis 13
Introduction 23
Guiding Principles for the Use of Digital Public Health
Technologies for Pandemic Response 25
1 Public Health Perspective 29
Types of Information Collected through Contact Tracing 29
How Contact Tracing Information Informs Public Health Action 31
Characteristics That Make Data Useful to
Public Health for Reducing Disease Transmission 32
2 Digital Technology and Contact Tracing 35
Characteristics of SARS-CoV-2 Relevant to Candidate Digital Solutions 35
Previously Existing Contact Tracing Technologies 36
Introduction of Novel Digital Contact Tracing Technologies 37
Relevant Differences between Manual and Digital Contact Tracing 41
vi Contents
3 Ethics of Designing and Using DCTT 43
Justifying the Use of DCTT Systems 44
Monitoring and Evaluating Technologies to Inform Policy and Practice 45
Public Trust and Public Attitudes 48
Designing Flexible Technology to Maximize Public Health Utility
While Respecting Other Values 50
Policy Positions to Advance Widespread Use of Digital Contact
Tracing Technologies 59
Disclosure and AuthorizationConsent 63
Promoting Equity and Fairness 69
Instituting Transparent Governance and Oversight 72
4 Legal Considerations 75
Data Privacy and Data Security Laws 77
Health Information Privacy 82
Labor and Employment Privacy Rights 84
Constitutional Privacy Rights 86
Consent 93
Anti-discrimination and Individual Freedom Laws 94
5 Recommendations 97
Public Health 97
Ethics 98
Legislative 102
Resources 103
US Government Response 103
Other Governmental and Nongovernmental Organizations 106
Digital Contact Tracing Experiences from Other Countries 112
Specific Digital ProductsApps 115
Polling 117
Popular Press 119
Commentaries 120
Academic Literature 121
Works Cited 127
Lead Authors and Contributors
vii
Lead Authors
Joseph Ali JD Assistant Professor Department of International Health Johns Hopkins Bloomberg School of Public Health Core Faculty amp Associate Director for Global Programs Johns Hopkins Berman Institute of Bioethics
Anne Barnhill PhD Core Faculty amp Research Scholar Johns Hopkins Berman Institute of Bioethics
Anita Cicero JD Deputy Director Johns Hopkins Center for Health Security Visiting Faculty Johns Hopkins Bloomberg School of Public Health
Katelyn Esmonde PhD Hecht-Levi Postdoctoral Fellow Johns Hopkins Berman Institute of Bioethics
Amelia Hood MA Research Program Coordinator Johns Hopkins Berman Insti-tute of Bioethics
Brian Hutler PhD JD Hecht-Levi Postdoctoral Fellow Johns Hopkins Berman Institute of Bioethics
Jeffrey Kahn PhD MPH Andreas C Dracopoulos Director Johns Hopkins Ber-man Institute of Bioethics
Alan Regenberg MBE Director of Outreach amp Research Support Associate Fac-ulty Johns Hopkins Berman Institute of Bioethics
Crystal Watson DrPH MPH Senior Scholar Johns Hopkins Center for Health Security Assistant Professor Department of Environmental Health and Engi-neering Johns Hopkins Bloomberg School of Public Health
Matthew Watson Senior Analyst Johns Hopkins Center for Health Security Senior Research Associate Department of Environmental Health and Engi-neering Johns Hopkins Bloomberg School of Public Health
Other Contributors
Robert Califf MD MACC Head of Clinical Policy and Strategy Verily and Goo-gle Health
Ruth Faden PhD MPH Philip Franklin Wagley Professor of Biomedical Ethics amp Founder Johns Hopkins Berman Institute of Bioethics
viii Lead Authors and Contributors
Divya Hosangadi MSPH Senior Analyst Johns Hopkins Center for Health Secu-rity Research Associate Department of Environmental Health and Engi-neering Johns Hopkins Bloomberg School of Public Health
Nancy Kass ScD Deputy Director for Public Health amp Phoebe R Berman Pro-fessor of Bioethics and Public Health Johns Hopkins Berman Institute of Bioethics
Alain Labrique PhD MHS MS Professor Department of International Health Johns Hopkins Bloomberg School of Public Health Director Johns Hopkins Global Health Initiative
Deven McGraw JD MPH LLM Chief Regulatory Officer CiitizenMichelle Mello JD PhD Professor of Law Stanford Law School Professor of
Health Research and Policy Stanford University School of MedicineMichael Parker BEd (Hons) MA PhD Director Wellcome Centre for Ethics and
Humanities Ethox Centre University of OxfordStephen Ruckman JD MSc MAR Senior Advisor to the President for Policy
Office of the President Johns Hopkins UniversityLainie Rutkow JD MPH PhD Senior Advisor to the President for National Cap-
ital Academic Strategy Office of the President Johns Hopkins UniversityJosh Sharfstein MD Vice Dean for Public Health Practice and Community
Engagement Professor of the Practice Johns Hopkins Bloomberg School of Public Health
Jeremy Sugarman MD MPH MA Deputy Director for Medicine Harvey M Meyerhoff Professor of Bioethics and Medicine Johns Hopkins Berman Institute of Bioethics Department of Medicine Johns Hopkins School of Medicine and Department of Health Policy and Management Johns Hop-kins Bloomberg School of Public Health
Eric Toner MD Senior Scholar Johns Hopkins Center for Health Security Senior Scientist Department of Environmental Health and Engineering Johns Hop-kins Bloomberg School of Public Health
Marc Trotochaud MSPH Analyst Johns Hopkins Center for Health Security Research Associate Department of Environmental Health and Engineering Johns Hopkins Bloomberg School of Public Health
Effy Vayena PhD Professor Health Ethics amp Policy Lab Department of Health Sciences amp Technology ETH Zurich
Tal Zarsky JSD LLM LLB Professor of Law University of Haifa Faculty of Law Visiting Scholar University of Pennsylvania Law School (2019ndash2020)
Preface
ix
Digital technologies are being developed and promoted to support the public health response to the COVID-19 pandemic with discussion and implementation planning in the United States by localities states institu-tions and employers Key decision makers and stakeholdersmdashincluding government officials institutional leaders employers digital technology developers and the publicmdashrequire clear and well-supported guidance to inform the deployment and use of these technologies as well as of the data they collect store and share While technology-based approaches are currently unable to provide solutions on their own experiences in other countries indicate that they could be used successfully in conjunc-tion with traditional and novel public health methods
This report reflects a rapid research and expert consensus group ef-fort led by the Berman Institute of Bioethics and the Center for Health Se-curity at Johns Hopkins University It draws on experts from both inside and outside Johns Hopkins in bioethics health security public health technology development engineering public policy and law The report highlights issues that must be addressed and provides recommendations for the use of digital technologies as part of contact tracing
The analysis offered here is focused on answering the following questions
bull Can digital contact tracing technologies (DCTT) be effective as part of public health responses to the pandemic and if so to what degree for which specific types of functions with what confidence and with what requirements
bull How can these technologies serve the interests of public health while respecting other individual and collective interests such as ensuring equitable distribution of benefits and burdens and limit-ing infringement on privacy and other civil liberties
x Preface
bull What are the ethical legal policy and governance guardrails cur-rently in place around such technologies and what else is needed
bull What additional guardrails are required to ensure that the goals of public health in using these technologies are achievable in ways that are ethically and legally sound
To answer these questions the report examines some core aspects of dig-ital technologies applied to contact tracing focusing on
bull the value of and basic methods for traditional public health surveil-lance and contact tracing
bull candidate technological products to enhance public health surveil-lance and contact tracing how they work and their comparative value for public health
bull core ethical legal and governance considerations and how they relate to relevant features of candidate technological solutions and
bull what is needed to move forward responsibly with the use of digital technology in support of public health surveillance acknowledg-ing gaps in our current understanding
The project involved in-depth analysis by a dedicated team of faculty postdoctoral fellows and research staff working over the course of only a few weeks but with great intensity drafting a report in collaboration with 26 total contributors writing commenting and revising through multiple drafts with the penultimate draft ldquopressure-testedrdquo by review and discus-sion at a virtual workshop of invited experts and stakeholders held on May 13 2020 and the final version completed on May 21 2020 The report and analysis builds on the excellent work of others in some parts of this territory while focusing on the gaps in analysis and areas that have not been sufficiently addressed The goal is to offer comprehensive guid-ance to relevant stakeholders to advance public health response during the COVID-19 pandemic Given the rapidly evolving territory into which DCTT is being introduced this report will by necessity be something of a living document updated as often as information dictates in order to continue to offer leading-edge analysis and guidance Versions will be noted in the digital and print editions
Acknowledgments
xi
Efforts like this project require teams and even small armies to be carried out successfully and this was no exception except that it was many fewer people working many more hours than could reasonably be expected of them From the initial kernel of an idea to the publication of this report in book form this project took just over a month total That seems impossi-ble even as I know it is accurate and it speaks to the incredible commit-ment hard work research skills and analytic acumen of our colleagues at Johns Hopkinsmdashthe core team are deservedly listed as lead authors of this report
None of this would have been possible without the supportmdashmoral and financialmdashand encouragement of Johns Hopkins University Presi-dent Ronald J Daniels who was the first to suggest the idea to me of taking on this topic He provided not only support and encouragement but the imprimatur of his office including help guidance and counsel from Prof Lainie Rutkow senior advisor to the president Lainie played a more integral role than that description captures reflected in part by her inclusion among the reportrsquos contributors but she deserves special acknowledgement for shepherding us through to the end
I mentioned that this was a team effort and every team requires an effective leader My colleague Prof Joseph Ali stepped into that role as we undertook the project and then he worked seven days a week along with the rest of the core research and writing team always unfailingly positive and deeply engaged in the work He along with Prof Anne Barnhill Alan Regenberg Amelia Hood and Drs Katelyn Esmonde Brian Hutler and Crystal Watson all deserve special thanks for doing so much in so little time all while working under the grinding social distancing restrictions of the 2020 pandemic That work was supported by Arnold amp Porter Kaye Scholer LLP with legal research and other assistancemdasha huge thanks to
xii Acknowledgments
them Finally the 16 contributing authors were incredibly generous with their time energy and insights all on ridiculously tight timelines and never a complaint or objection
The project benefited greatly from a number of experts who provided written feedback on drafts and who attended the virtual workshop to test our recommendations including Miles Stewart Rob Nichols Smisha Aagarwal Karl Steiner Anupam Joshi Charles Scheeler Ford Rowan and Jay Wagley
Last the fact that this report appears in published book form by Johns Hopkins University Press is another minor miracle from manu-script to printed book in under a week Thanks to JHUP Director Barbara Kline Pope and her team for being willing to take on the challenge and for the incredible focused effort it required
my heartfelt thanks andappreciation to you all
Jeffrey Kahn
Acronyms and Abbreviations
xiii
ADA Americans with Disabilities ActBLE Bluetooth Low EnergyCalOPPA California Online Privacy Protection ActCBP Customs and Border ProtectionCCPA California Consumer Privacy ActCDC Centers for Disease Control and PreventionCLOUD Act Clarifying Lawful Overseas Use of Data ActCOPPA Childrenrsquos Online Privacy Protection ActCOV+ confirmed positive SARS-CoV-2 test resultCOVID-19 coronavirus disease 2019CPNI customer proprietary network informationCSLI cell-site location informationDCTT digital contact tracing technology and closely related
digital health productsECPA Electronic Communications Privacy ActEEOC Equal Employment Opportunity CommissionE-SIGN Electronic Signatures in Global and National
Commerce ActEU European UnionFCC Federal Communications CommissionFTC Federal Trade CommissionFTCA Federal Trade Commission ActGIS geographic information systemGPS global positioning systemHHS US Department of Health and Human ServicesHIPAA Health Insurance Portability and Accountability ActHIV human immunodeficiency virusICU intensive care unit
xiv Acronyms and Abbreviations
IRB institutional review boardJHU Johns Hopkins UniversityOCR Office for Civil Rights US Department of Health amp
Human ServicesOSHA Occupational Safety and Health AdministrationPHI protected health informationPII personally identifiable informationPPE personal protective equipmentPPPT privacy-preserving proximity trackingQR Code quick response codeRFRA Religious Freedom Restoration ActRLUIPA Religious Land Use and Institutionalized Persons ActSARS-CoV-2 severe acute respiratory syndrome coronavirus 2SCA Stored Communications ActSTI sexually transmitted infection
DIGITAL CONTACT TRACING FOR PANDEMIC RESPONSE
11
Summary
Introduction
Public health professionals around the world are working tirelessly to respond to the COVID-19 pandemic using tried-and-tested public health methods for infectious disease surveillance and control These traditional methods are essential to the global COVID-19 response To complement these actions and potentially augment the speed and efficacy of the pub-lic health workforce digital technologies are being harnessed Given the scale of the pandemic significant efforts are being undertaken to develop and leverage public-facing and health-system-supportive technology solu-tions including smartphone apps and other digital tools that may aid public health surveillance and contact tracing
Digital contact tracing technology and closely related digital health products (together DCTT) have been used in several countries as part of broader disease surveillance and containment strategies In the United States DCTT has been proposed as an integral part of some plans to ldquoreopenrdquo the country (Allen et al 2020 Hart et al 2020 Simpson and Conner 2020) It is almost certain that these and related technologies will become part of not only the COVID-19 response but also the larger toolbox for future public health communicable disease prevention and control
These technologies have significant promise They also raise import-ant ethical legal and governance challenges that require comprehensive analysis in order to support decision-making Government officials pub-lic health leaders leaders of institutions employers digital technology de-velopers and the public all must be adequately informed in order to make
2 Digital Contact Tracing for Pandemic Response
responsible choices Johns Hopkins University recognized the importance of helping to guide this process It organized an expert group with mem-bers from inside and outside of Hopkins and led by its Berman Institute of Bioethics in collaboration with the Center for Health Security Its charge was to examine the ethics law policy and public health implications of using digital technologies as part of pandemic response and to develop guidance including a framework and actionable recommendations for governmental and institutional decision makers
Overall this expert group urges a stepwise approach that prioritizes align-ment of technology with public health needs and public values building choice into design architecture and capturing real-world results and impacts to allow adjustments as required Further we urge an approach that recog-nizes that there are complicated issues to resolve for governments insti-tutions and businesses and that introduction of DCTT must include public engagement and ongoing assessments to improve both performance and adoption
Specific recommendations include the following
bull There is no ldquoone size fits allrdquo approach to DCTT Technology design should not be static but it should be capable of evolving depending upon local conditions new evidence and changing preferences and priorities
bull Technology companies alone should not control the terms condi-tions or capabilities of DCTT nor should they presume to know what is acceptable to members of the public
bull DCTT should be designed to have a base set of features that pro-tect privacy with layers of additional capabilities that users may choose to activate An initial default should be that user location data are not shared but users should be provided with easy mech-anisms and prompts to allow for opting-in to this capability with encouragement to the public if it is shown to be critical to achieving public health goals
bull Data collected through DCTT should be made available to public health professionals and to researchers in de-identified form to support population-level epidemiologic analyses
Summary 3
bull Those who authorize use of DCTT within a particular jurisdic-tion or institution should continuously and systematically monitor the technologyrsquos performance in that context This should include monitoring for effectiveness and benefit monitoring for harms and monitoring for the fair distribution of both benefits and harms
bull Governments should not require mandatory use of DCTT given uncertainty about potential burdens and benefits Additional tech-nology user and real-world testing is needed
Through in-depth analysis and recommendations this report seeks to guide decision-making and enhance understanding of
bull the value of and basic methods for traditional public health surveil-lance and contact tracing
bull candidate technological products to enhance public health surveil-lance and contact tracing and their comparative value for public health
bull core ethical legal policy and governance considerations and how they relate to relevant features of candidate technological solutions and
bull what is needed to move forward responsibly with the use of digital technology in support of public health surveillance acknowledg-ing gaps in our current understanding
The full set of recommendations are intended to (1) support effective and informed adoption of DCTT (2) encourage design of flexible technol-ogies that maximize public health utility while respecting other values (3) establish meaningful processes for user disclosure and authorization (consent) (4) promote equity and fairness in the uses of DCTT and (5) foster transparent governance and oversight
DCTT Features Functions and Potential Applications
Digital contact tracing technologies and platforms can be roughly catego-rized into three broad approaches along a spectrum of potential policies and methods a maximal approach (typified by the South Korean govern-
4 Digital Contact Tracing for Pandemic Response
mentrsquos centralized and triangulated data collection (M S Kim 2020)) a minimal approach (typified by the AppleGoogle decentralized priva-cy-preserving proximity tracking (PPPT) and contact notification (Apple and Google nd)) and a diverse range of middle-ground approaches that aim to augment manual contact tracing with the collection of digital data that can be shared with public health authorities
Minimal approaches such as the AppleGoogle PPPT use Bluetooth Low Energy (BLE) ldquohandshakesrdquo that record close contact between mo-bile phone users but do not register the location in which the contact hap-pened In most architectures these proximity data are stored in the usersrsquo phones as anonymized ldquobeaconsrdquo that cannot be used to re-identify the users directly If a user with a PPPT app installed on their phone tests pos-itive and enters test results into their app those who have been in contact with them can be notified by the app This ldquoexposure notificationrdquo can be automatic or at the discretion of the COV+ person depending on the app design If notified a user who has been in contact with a COV+ individ-ual would receive a push notification alerting them to possible exposure (which may be timestamped) but with no other identifying information
The most prevalent middle-ground approach in the US context in-volves the collection and storage of personal datamdashincluding identifying information and location datamdashon the userrsquos phone These decentralized but personally identifiable data can then be voluntarily shared with pub-lic health officials if the user tests positive for SARS-CoV-2 (severe acute respiratory syndrome coronavirus 2) For example a team at the Mas-sachusetts Institute of Technology (MIT) has developed an app called Private Kit Safe Paths (MIT nd) that stores usersrsquo location data on their phone for 28 days If a user tests positive she can voluntarily upload her location data to a website that is accessible only to public health officials Officials can then analyze these personally identifiable data and subse-quently broadcast redacted and de-identified data to other users Healthy users would have access to these redacted location data of COV+ users but their own data would not leave their phones At a minimum the storage of user location data can function as a ldquomemory aidrdquo if the user tests positive but releasing the data to public health authorities may help to analyze the spread of SARS-CoV-2 and alert individuals or groups that have been in contact with COV+ patients
Summary 5
The US Centers for Disease Control and Prevention (CDC) has pub-lished preliminary criteria for evaluating capabilities and attributes of DCTT (CDC 2020e) These and other resources suggest that a compre-hensive assessment of DCTT and its potential to advance the publicrsquos health will require careful consideration of numerous interconnected fac-tors that interact in complex ways and must be navigated within the chal-lenging contexts of uncertainty and urgent need (Figure 1) These include
bull scientific and epidemiological understanding of SARS-CoV-2 transmission and infection
bull public health needs for combating the outbreak
bull technological capabilities of DCTT
bull performance of DCTT applications
bull ethical values and principles
bull characteristics of public adoption and acceptance and
bull legal issues and landscape
FIGURE 1 Interrelating Factors That Frame Responsible Development of Digital Contact
Tracing Technology
6 Digital Contact Tracing for Pandemic Response
The primary objectives for use of DCTT during the COVID-19 pandemic must be to reduce illness and death and facilitate public health efforts to reduce transmission of the virus These objectives fall under a broader overall goal of contributing to societal well-being during the pandemic It is not yet known whether and how much DCTT can contribute to these primary objectives nor whether it will be able to contribute without gen-erating new burdens or even harms such as incorrect warnings or ldquonoiserdquo that detract from the work of manual contact tracing
The process of identifying acceptable technology designs and uses is complex given the interplay among the factors Our analysis reveals that there is no ldquoone size fits allrdquo approach to DCTT There is variability across the United States with respect to SARS-CoV-2 prevalence and in-fection rates public health capacity public attitudes toward DCTT and acceptability of various potential features Moreover our understanding of SARS-CoV-2 and DCTT is evolving public health response needs and capabilities are changing and public attitudes are shifting Different tech-nologies used in different ways may be appropriate to achieve slightly dif-ferent public health goals in different localities and at different points in the pandemic A tiered and phased approach to technology development should be facilitated by law and policy prioritizing underlying interoper-ability while permitting user choices now and for the future
Given the complexity of the terrain as a first step those developing or considering widespread use of DCTT as part of pandemic response should be guided by the following principles and related actions (see box) These principles are meant to apply to DCTT as well as other dig-ital technologies used in novel ways during pandemic response
These principles make clear that in order to maximize the public good from use of DCTT public health needs and technological capabili-ties must be carefully aligned Government officials public health leaders leaders of other institutions employers digital technology developers and the public are all key stakeholders that must be informed and en-gaged in order to enable the most successful and ethically acceptable uses of DCTT
Guiding Principles for the Use of Digital Public Health Technologies for Pandemic Response
Transparency and public engagement are essential to an inclusive digital public health response
bull Government public health and digital technology leaders must engage
effectively with the public and other stakeholders to communicate the util-
ity importance oversight and limitations of relevant digital technologies
including their implications for individualsrsquo privacy and civil liberties
bull Transparency at all levels is essential for maintaining public trust and
confidence
bull To the extent possible digital public health responses should reflect the
range of values that are important to individuals including advancing the
health and well-being of the community as a whole
bull Decision makers should recognize the sacrifices that some people may be
willing to make during a pandemic in order to advance public health goals
Acceptance by some of particular monitoring capabilities should not be read
as a willingness to extend these methods to other problems or uses
Digital public health responses must represent the least infringement of civil liberties necessary to accomplish the public health goals
bull If preferred digital public health strategies infringe on privacy and other civil
liberties the infringements must be sufficiently justified by the circum-
stances of the pandemic offset by ample anticipated public benefit and
considered relative to infringements associated with other possible strate-
gies such as mass physical distancing
bull Only those data that are necessary and relevant for the stated public health
purposes should be collected Identifiable data should be stored in a secure
manner and only for the period of time that the public health purposes
require
bull Adopted technologies should not be used in ways that subject communities
to discrimination or surveillance for nonndashpublic health reasons
bull Respect for individual autonomy requires that users are sufficiently informed
of the public health goals of the technology and the extent to which those
goals are being met
Use of digital public health technologies and data must be guided by best available evidence
bull Decisions to deploy digital public health technologies should be based on a
careful assessment of the uses and limitations of any proposed technology
taking into account the best available evidence
bull Those who deploy digital public health technologies should continuously and
systematically monitor their performance as well as any evidence that is being
generated in other contexts about the selected technological solution and about
other competing technologies
bull Unintended consequencesmdashincluding those that might impact public health
goals core values and interests of the public and unfair advantage or disadvan-
tagemdashshould be carefully monitored and addressed as necessary
Responsible use of digital public health technology requires meaningful governance and accountability
bull Systems of governance must be trustworthy and well informed They must be
reviewed and adjusted as circumstances and evidence change or as unintended
effects are identified
bull Trusted representatives who are capable of developing and implementing uni-
form and fair standards for adopting and utilizing underlying digital technology
must be identified
bull Understandable transparent and publicly accessible rules must guide the
collection access control use storage and combination of data by government
authorities public and private institutions and other parties such as public
health researchers
bull Oversight accountability and consequences for abuse or misuse of these data
must be explicit and enforceable
The deployment of digital public health technology must be rooted in a commitment to equity
bull Digital public health technologies should be deployed in a manner that does
not propagate preexisting patterns of unfair disadvantage or further distribute
harms and risks unfairly throughout the population
bull To the extent possible digital public health technologies should be designed to
rectify existing inequities
bull Oversight mechanisms must be in place to ensure that the improved public
health outcomes are equitable and to detect and correct any unforeseen resul-
tant injustices attributable to the technology or that can be addressed using the
technology
bull The incentives and disincentives for adopting new technology must be equita-
ble not exploitative and aligned with effective use of the technology
bull Disparity-driven technology gaps should be explicitly recognized To the extent
possible provisions should be made to address the digital divide
Summary 9
Summary of Recommendations
The guidance document makes a number of recommendations related to (1) supporting effective and informed adoption of DCTT (2) designing flexible technologies to maximize public health utility while respecting other values (3) establishing meaningful processes for user disclosure and authorizationconsent (4) promoting equity and fairness in application of DCTT and (5) instituting transparent governance and oversight Here we provide a summary of recommendations
Supporting Effective and Informed Adoption
bull Those who authorize use of DCTT within a particular jurisdic-tion or institution should continuously and systematically monitor the technologyrsquos performance in that context This should include monitoring for effectiveness and benefit monitoring for harms and monitoring for the fair distribution of both benefits and harms They should also monitor evidence that is being generated in other contexts about their selected technological solution and about other competing technologies
bull Data collected through DCTT should be made available to public health professionals and to researchers in de-identified form to support population-level epidemiologic analysis
bull Data should be available to users that would permit them to further investigate their personal risk with public health officials or other health workers to add a layer of protection against unnecessary quarantine
bull Technologies or apps may produce some false negatives or false positives but they should be accurate enough that public health authorities feel confident that they support and donrsquot detract from contact tracing efforts
bull Trusted leaders should be enlisted to communicate effectively with the public about DCTT and encourage its use should the technol-ogy demonstrate some potential The limits of knowledge regard-ing effectiveness should also be explained along with what will be done to improve technological capabilities as understanding evolves
10 Digital Contact Tracing for Pandemic Response
bull Incentives can be a useful complement to encouragements how-ever any incentives for users to install and use DCTT must be equitable should not be coercive and should align with effective use of the technology
bull DCTT use should not be mandated at this time given uncertainty about potential harms and benefits Additional technology user and real-world testing is needed
Designing Flexible Technology to Maximize
Public Health Utility While Respecting Other Values
bull Technology companies should not alone control the terms condi-tions or capabilities of DCTT nor should they presume to know what is acceptable to members of the public
bull A ldquovalues in designrdquo approach to development of DCTT should be adopted (Flanagan Howe and Nissenbaum 2008 Knobel and Bowker 2011) Robust public and user engagement activities should be pursued to identify and incorporate to the extent pos-sible a range of values into the design of the technology These values may include privacy but also autonomy efficiency equity or others Technology design should reflect an appropriate balance and prioritization of identified values
bull Technology design should not be static but should be capable of evolving depending upon local conditions new evidence and changing preferences and priorities
bull DCTT should be designed to have a base set of features that pro-tect privacy with layers of additional capabilities that users may choose to activate An initial default should be that user location data are not shared but users should be provided with easy mech-anisms and prompts to allow for opting-in to this capability with encouragement to the public if it is shown to be critical to achieving public health goals
Establishing Meaningful Processes for
User Disclosure and Authorization (Consent)
bull A clear and concise module consisting of basic disclosure and vol-untary authorization should be developed to accompany DCTT
Summary 11
This module should not take the form of ldquoclickwraprdquo terms of service or end-user agreements but rather provide only essential information necessary for an individual to make a decision More detailed disclosures (such as FAQs in plain language) should be made easily accessible to those who wish to learn more with no hidden surprises
bull An opt-in approach to authorization should be instituted to ac-company initial DCTT rollout The feasibility and value of opt-out approaches should continue to be evaluated informed by what is technologically possible what local assessments of benefits and harms of the technology reveal over time and our evolving un-derstanding of the degree to which an opt-out approach is likely to increase or decrease utilization among different populations Opt-out approaches should not be precluded
Promoting Equity and Fairness in Application of DCTT
bull States localities and institutions that recommend widespread use of DCTT should provide technology (eg mobile phones Blue-tooth devices) and free data packages to those who desire but lack access to these devices
bull If there are lower rates of adoption of DCTT systems in some iden-tifiable communities public health authorities should find ways to compensate For example directing more non-DCTT resources and efforts toward those communities to meet specific needs that are elsewhere being supported by technology
bull If maps are generated based on DCTT to provide the public with the locations that COV+ individuals have visited steps must be taken to minimize the stigma and potential financial losses that could result from a location being identified as a hotspot
Instituting Transparent Governance and Oversight
bull Digital surveillance oversight committees should be established expeditiously with diverse and qualified membership to provide ethical and regulatory review prior to and concurrent with wide-spread use of a DCTT system
bull Understandable and publicly accessible rules must guide the col-
12 Digital Contact Tracing for Pandemic Response
lection access control use storage and combination of data by government authorities public and private institutions and other parties such as public health researchers
bull Only those data that are necessary and relevant for the public health response to COVID-19 should be collected and used
bull Identifiable data should be kept only for the period of time needed for the public health response to COVID-19
bull Identifiable data collected as part of this response should not be shared with anyone other than the relevant public health authorities without additional specific informed consent of individual users
bull Before a government or institution adopts a digital contact tracing program they should state the conditions under which the digital contact tracing program will be terminated
bull Future use of DCTT to advance public health or other efforts (eg use in seasonal flu surveillance) would require independent justification DCTT designed for public health use should not be used by law or immigration enforcement
bull The principles offered in this guidance document apply both during and following the COVID-19 pandemic
Legislative Recommendations
bull The United States Congress should enact new legislation specif-ically tailored to facilitate the use of DCTT as part of the public health response to COVID-19 while also protecting user privacy and ensuring data security
bull Congress should require DCTT developers to disclose to users in clear language the nature of the information that would be col-lected how it would be collected how it would be stored and for what purposes it may be used
bull While the rollout of DCTT should initially employ an opt-in au-thorization approach the feasibility acceptability and value of opt-out approaches should continue to be evaluated As such opt-out approaches to consent should not be precluded by legislation
Summary 13
bull Congress should prohibit the commercial use of data collected for COVID-19 response by DCTT
bull Congress should prohibit discrimination on the basis of data col-lected by DCTT
bull If Congress is unable to enact suitable legislation state legislatures should work toward enacting similar laws for their jurisdictions A ldquomodelrdquo state law should be rapidly developed to facilitate na-tionwide adoption of an appropriate law and uniformity of legal requirements
Summary of Analysis
Supporting Effective and Informed Adoption
The COVID-19 pandemic and the physical distancing efforts imple-mented to slow the rate of transmission have caused severe harm to indi-viduals communities and our society To protect the public good going forward we need a robust public health response that reduces the spread of SARS-CoV-2 and does so in a way that allows economic recovery to occur and to be sustained We also need to design and manage this public health response so as to minimize harms to individuals and society to distribute benefits and burdens equitably across the population and to avoid misuses of the technologies and the data they collect
To reduce the spread of SARS-CoV-2 chains of transmission need to be broken To do this people who have been exposed to SARS-CoV-2 or potentially exposed need to be identified as comprehensively and as quickly as possible so they can quarantine themselves and avoid infecting others This is the job of manual contact tracing by public health authori-ties in which people infected or presumptively infected with SARS-CoV-2 are interviewed and asked about their movements and interactions in-cluding where they work and shop how they travel with whom theyrsquove had contact and the nature of that contact (eg where the contact took place) Their contacts are then interviewed and potentially asked to quar-antine seek testing and take other protective measures if the contact is sufficiently high risk
14 Digital Contact Tracing for Pandemic Response
The hope is that DCTT can augment traditional contact tracing ef-forts either by working alongside and independently of manual contact tracing or by being integrated into manual contact tracing efforts in a way that makes these efforts faster more thorough and more efficient
Data suggest that a substantial proportion of transmissionsmdashper-haps as high as 50mdashoccur between individuals who are not symptom-atic and that transmission may occur as early as 3 days before onset of symptoms (WHO 2020) Because asymptomatic spread of SARS-CoV-2 appears to be a significant source of infection we need to identify po-tentially infected people before they show symptoms thus speed is of the essence This is one benefit of using DCTT potential contacts can be identified instantaneously notified quickly and asked to quarantine as soon as possible
Another benefit is identifying contacts who manual contact tracing methods may miss either because COV+ people do not remember all the places theyrsquove been or cannot identify all the people theyrsquove had contact with This is especially relevant given the long period of infectivity of SARS-CoV-2 which begins before people are symptomatic and aware they are infected (Ferretti et al 2020) If DCTT were designed to have optional location-monitoring capabilities this critical challenge could be mitigated even further For example location data might reveal that a COV+ person was at a restaurant at an exact time and date which could be followed up by contact tracers who could alert the public or use other measures to reach those who were also present in the restaurant at the same time In other disease contexts geolocation data have demonstrated some potential to support epidemiology and disease surveillance (see Fur-lanello et al 2002 Dredze et al 2013 Eckhoff and Tatem 2015 Fraccaro et al 2019) with technical cautions regarding accuracy and the like (Beu-kenhorst et al 2017)
One role for DCTT is to work alongside manual contact tracing but independently of it Individuals would download proximity tracing or exposure notification apps use them receive alerts if theyrsquove had a poten-tial contact with another user who is COV+ or presumptively COV+ and voluntarily self-quarantine without having contact with public health au-thorities or giving them data that feeds into public health contact tracing efforts It is possible that this would help to break chains of transmission and reduce the spread of SARS-CoV-2 though at this point these benefits
Summary 15
are speculative It is also possible that such exposure notifications will result in high rates of false positives
Another possible role is for DCTT to be integrated into manual con-tact tracing efforts When potential contacts are identified by DCTT they are connected to public health authorities who can then follow up with them There are different forms this could take and different kinds and amounts of data about contacts public health authorities could receive from DCTT On one end of the spectrum of reporting public health authorities would not receive individualsrsquo names or contact information only anonymous data The fullest version of reporting would securely send to public health authorities the names contact information such as address and phone number and other data about contacts that DCTT collected including data about their location and movement history
It is uncertain whether providing public health authorities with vol-umes of information on cases and contacts from DCTT will be useful in practice As mentioned above providing public health authorities with location data on cases and contacts collected by DCTT may help con-tact tracers to find and notify additional contacts However at present providing public health authorities with large amounts of data will be useful only if there is sufficient capacity to follow up on these data In addition there is a risk of low-quality data from DCTT flooding the system leading to investigation of false case contacts identified by DCTT and distracting from other important efforts Whether and to what extent data from DCTT will benefit contact tracing efforts is unknown pointing again to the importance of continuously collecting high-quality evidence about DCTT
Designing Flexible Technology to Maximize
Public Health Utility While Respecting Other Values
Use of DCTT is essentially an experiment as we have insufficient infor-mation about the performance of different DCTT and their efficacy In the face of this uncertainty how should DCTT be designed and how should its use be managed
Many efforts to advance DCTT in the United States and elsewhere have emphasized the importance of ldquoprivacy by designrdquo that is building privacy and security protections into the design of technology rather than counting on responsible use alone (Cavoukian 2010) As noted above
16 Digital Contact Tracing for Pandemic Response
some major technology companies have signaled this position through de-velopment of PPPT systems that embed features such as decentralization de-identified information user anonymity bans on collection of location data and minimal reliance on or integration of public health authorities or other government actors Many of these features have also been em-braced early by advocacy organizations (Crocker Opsahl and Cyphers 2020 Electronic Privacy Information Center 2020 Kahn Gilmor 2020) and in an open letter (ldquoJoint Statement on Contact Tracingrdquo 2020) from nearly 300 researchers These same actors have emphasized that use of DCTT should be fully voluntary
Although privacy is a key value individuals and communities may also value efficiency equity liberty autonomy economic well-being com-panionship patriotism or solidarity among other values People may accept more significant encroachments on privacy now if this ultimately results in realizing other values (such as companionship) that are of equal or greater importance to those individuals Rather than centering pri-vacy alone in design a different orientation is needed at this moment that of ldquovalues in designrdquo which incorporates a broader range of values into technology (Flanagan Howe and Nissenbaum 2008 Knobel and Bowker 2011) For example some users might wish to express autonomy solidarity or patriotism through DCTT by sharing their location history with public health professionals in order to advance the public health re-sponse increase system efficiencies (eg by contributing information that can lead to better data processing) and reduce the burden on essential workers At the same time there is value in further advancing autonomy by designing technology to allow individuals some control over what data about them are collected and shared
DCTT should be designed to have a base set of features that protect privacy and strive for interoperability but also should include other op-tional capabilities This could be achieved by designing DCTT to have a default that can be modified for example an initial setting could be that usersrsquo location data are not shared with public health authorities but us-ers may opt-in to this feature Such an opt-in approach is likely consistent with existing federal privacy laws
Designing DCTT this way gives users the flexibility to decide how to use the technology and how to engage with public health authori-ties consistent with their values and trade-offs they are willing to make This flexibility could also allow for more real-world evaluation of how
Summary 17
different users experience different features of DCTT in different loca-tions Technology design should not be static but it should be capable of evolving depending upon local conditions new evidence and changing preferences and priorities
DCTT developers must comply with a number of federal privacy laws These privacy laws generally permit the collection storage and use of personal information so long as the user provides meaningful consent Privacy law in the United States is generally sector-specific and limited in scope resulting in a patchwork of protections that differ significantly depending on the entity that collects the data and the type of data col-lected Given the complexity of existing federal privacy law and the need to further strengthen public trust in DCTT it would be beneficial for Congress to enact new privacy legislation that is specifically tailored to the use of DCTT in response to COVID-19 Such COVID-specific legisla-tion should be sensitive to the full range of values and recommendations described above
In short designing ldquomiddle-groundrdquo DCTT for flexible use may pro-vide the most adaptable and thus most robust public health responsemdashrespecting privacy and individual autonomy by allowing users to use DCTT in ways that express their own values
Public Acceptance of DCTT
While some groups have maintained that only PPPT-like minimal systems will be widely adopted because only they will earn and maintain public trust (Simpson and Conner 2020) there is insufficient evidence that pub-lic trust would be threatened by a DCTT system that has the capacity to securely collect location data integrate public health authorities and en-able voluntary sharing of certain user data (eg location data) with those authorities More research including through deliberative engagement sessions is needed to better understand how differences in the features and functionality of DCTT (such as optional sharing of geolocation data) influence trust and peoplersquos willingness to use DCTT Technology com-panies should not alone control the terms conditions and capabilities of DCTT nor should they presume to know what is acceptable to members of the public
Significant concerns have also been expressed by privacy advocates (Guariglia 2020) and in the popular press (Giglio 2020) about ldquosurveil-lance creeprdquomdashthat is a belief that state or corporate actors will use new
18 Digital Contact Tracing for Pandemic Response
surveillance technologies capacities and permissions well beyond the purposes for which they were initially justified to the public and beyond the time when they are useful for the COVID-19 pandemic Surveillance creep is a serious concern and should be carefully guarded against how-ever the possibility of surveillance creep is not a sufficient reason to limit development of DCTT to minimal systems Instead protections should be put in place to ensure that only those data that are necessary and relevant for the public health purposes at hand are collected and used and data should be kept only for the period of time needed for those pub-lic health purposes For this reason we would support COVID-specific legislation that would impose strict limits on the use of DCTT data for nonndashpublic health purposes
Finally the use of DCTT during the current pandemic should not set a precedent for future public health use (eg use in seasonal flu surveil-lance efforts) Future use would require independent justification Further use of DCTT in other contexts (eg by law enforcement or immigration enforcement) is presumptively unethical
Encouraging Adoption of DCTT
Researchers have estimated perhaps conservatively that DCTT use by 80 of smartphone owners or 56 of the population overall will be needed to suppress the epidemic (Hinch et al 2020) These estimates also highlight that some decrease in transmission would be realizable even with lower rates of technology adoption
In the United States many advocates and researchers have argued that use of DCTT must be fully voluntary However experience from other countries suggests that when use of a digital contact tracing app is voluntary only a minority of the population will download it Instead of making use fully voluntary and initiated by users there are ways that DCTT could be put into use without usersrsquo voluntary choice For exam-ple use of an app could be mandated as a precondition for returning to work or school or even further to control entry into a facility or trans-portation (such as airplanes) through scanning of a QR code to demon-strate personal exposure levels (Gan and Culver 2020)
While these approaches are hard to imagine in the United States some have argued that mandatory use of DCTT could be ethical If man-dates increase adoption of DCTT and improve the public health response
Summary 19
this would reduce the likelihood of lockdowns which are harmful and a severe limitation of individual liberty applied on a mass scale On the other hand mandated use of DCTT systems may not be effective People may not adhere to the mandate by simply leaving their phone at home Perhaps more important should the technology not deliver the hoped-for benefits having mandated the use of an unproven technology could result in a loss of public trust in the technology in the entity instituting the mandate and in the larger public health response potentially lead-ing to noncompliance with public health recommendations more broadly (Bernstein et al 2019)
Any decision maker considering mandatory use including govern-ment officials institutional leaders and employers must convincingly address a number of considerations Particularly important is the need to identify reliable evidence that the DCTT would be effective and to ensure that the burdens and benefits of use are equitable and justifiable At this time mandated use of DCTT by states or institutions is not jus-tifiable given uncertainty about potential harms and benefits Additional technology user and real-world testing is needed before mandatory use should be considered
As with any public health effort the amount of evidence that must be offered to illustrate that the intervention or program can achieve its aims and the degree to which people should be able to exercise choice in their participation should be in proportion to the anticipated bur-dens of the intervention or program For example the permissibility of mandating use of DCTT by the public depends on factors such as the sensitivity of the data that are collected the extent to which public health is integrated within the DCTT system and what actions are taken in response to confirmed virus exposure or being identified as COV+ (eg forced quarantine) The more burdens that are placed on individualsmdashfor example whether people are ordered into quarantine if they have been exposed to the virus or if there are limited social supports for those in quarantinemdashthe greater the demand should be on the performance of the DCTT system
Perhaps the most effective way to generate widespread US adoption of DCTT will be to offer incentives for its use in other contexts gener-ally speaking small incentives have been shown to lead to an increase in desired outcome (Singer and Ye 2013 Lee et al 2014) Given the impor-
20 Digital Contact Tracing for Pandemic Response
tance of widespread use modest incentives ought to be considered if and when there is sufficient evidence of the utility of DCTT so long as those incentives are not mandates in disguise Another ldquofirst linerdquo approach to increasing use of DCTT is for trusted community leaders public figures health care professionals and other respected individuals to communicate with the public and their communities about DCTT and to encourage its use through public engagement campaigns if and when the technology demonstrates sufficient potential
Establishing Meaningful Processes for
User Disclosure and Authorization (Consent)
Any effort to roll out DCTT should ensure that users have a meaning-ful opportunity to review and understand information about the specific technology and its uses Moreover given the importance of public trust and the current crisis of public trust in governments and technology com-panies handling private digital information there is a strong ethics argu-ment for requiring consent from individual users We recommend a care-fully crafted version of what is sometimes called simple consent which consists of basic disclosure and voluntary agreement or authorization This disclosure should include information about the purposes of the technology the userrsquos options for collecting and sharing data purposes for which data can be used and any known risks among other informa-tion This information should be presented in an accessible format on any DCTT app and more detailed disclosures should be readily accessible for those who wish to review them
Through an opt-in mechanism such as clicking a button to signal agreement users should be able to indicate their intention to use a DCTT The opt-in approach is consistent with mechanisms for agreement to use other downloaded applications An opt-in approach should be part of the initial introduction of DCTT given the novelty of the technology and its uses and the need to build trust and confidence in the system Successes of opt-out approaches in other areas suggest that the feasibility and value of an opt-out approach to DCTT should be carefully evaluated particularly in conjunction with assessment of whether public health goals are being met (Rithalia et al 2009) Such assessments should be informed by what is technologically possible by local data regarding benefits and harms of the technology and by evolving understanding of the degree to which
Summary 21
an opt-out approach is likely to increase or decrease utilization among different populations
Promoting Equity and Fairness in Application of DCTT
Digital contact tracing technology should be designed and used in ways that as far as possible promote an equitable distribution of benefits and burdens DCTT should be deployed in a manner that does not propagate preexisting patterns of unfair disadvantage or distribute harms and risks unfairly throughout the population It is well known that some commu-nities have lower rates of technology and data access and therefore may benefit less from use of DCTT unless steps are taken to address these digital disparities Additionally should use of DCTT be made a require-ment for entry into a workplace into a school or onto transportation then those who currently do not possess the required technology must not be unfairly burdened through lack of access In order to mitigate this states localities and institutions that recommend widespread use of DCTT should provide technology (eg mobile phones Bluetooth de-vices) and free data packages to members of the community who desire but lack access to these devices
Some populations may also experience greater harm and greater fear of harm from having their data collected For example some groups such as African Americans Hispanic Americans Muslim Americans and undocumented immigrants have more reasonable fear of their data being handed over to law or immigration enforcement and some groups have lower levels of trust in public health due to past injustices (Auxier et al 2019 CSM 2017 Pew Research Center 2017 Rodrigues et al 2018) This further substantiates the need to limit use of any data gathered by DCTT to its public health purpose
Instituting Transparent Governance and Oversight
DCTT must be developed with an eye toward both present and future implications We are rapidly gaining knowledge about SARS-CoV-2 and COVID-19 but still have essential gaps in our understanding In the United States public health responses including DCTT will generally be developed and coordinated by individual states regional consortia (Res-ton Sgueglia and Mossburg 2020) and associations Good governance in this context requires transparency and the creation of oversight bodies
22 Digital Contact Tracing for Pandemic Response
with the appropriate expertise and representation to allow nimble and effective responses while serving as trusted representatives
In order to address the range of ethics and governance concerns that relate to the design and use of DCTT we recommend that digital surveil-lance oversight committees be established perhaps at a state level and with a platform for national coordination These committees can provide ethics and regulatory review prior to and concurrent with widespread use of DCTT The committees should be composed of a diverse group of experts capable of evaluating the quality of a DCTT system locally including members of communities that experience higher rates of digital disparity
When assessing the design and use of digital contact tracing systems these committees (and the public more widely) should consider not only the risks and benefits accrued during the COVID-19 pandemic but also implications for the future How can we navigate safe use of these tech-nologies in a way that preserves public trust in them and enables the possibility of future beneficial use
As a start it should be emphasized that the principles offered in this and other guidance documents do not apply only during the pandemic Future efforts to advance DCTT capabilities during quieter times should make every effort to follow them
23
Introduction
Public health professionals around the world are working tirelessly to respond to the COVID-19 pandemic using tried-and-tested public health methods for infectious disease surveillance and control These traditional methods are essential to the global COVID-19 response To complement these actions and potentially augment the speed and efficacy of the public health workforce digital technologies are being harnessed Given the scale of the pandemic significant efforts are being undertaken to develop and leverage public-facing and health-system-supportive tech-nology solutions including smartphone apps and other digital tools that may aid public health surveillance and contact tracing
Digital contact tracing technology and closely related digital health products (hereafter DCTT) have been used in several countries as part of broader disease surveillance and containment strategies Globally many digital COVID-19 contact tracing strategies have already emerged in response to the pandemic This is not surprising given the ubiquity of mobile phones and other digital devices around the world (ldquoMobile Cellular Subscriptions (per 100 People)rdquo 2018) experiences developed during prior outbreaks and pandemics and the pre-COVID-19 momen-tum behind using digital technologies to support individual and health system capabilities (WHO 2017 Mathews et al 2019 Aiello Renson and Zivich 2020 Mahmood et al 2020) In the United States DCTT has been proposed as an integral part of some plans to ldquoreopenrdquo the country (Allen et al 2020 Hart et al 2020 Simpson and Conner 2020) It is al-most certain that these and related technologies will become part of not only the COVID-19 response but also the larger toolbox for future public health communicable disease prevention and control
24 Digital Contact Tracing for Pandemic Response
While novel public health surveillance technologies such as DCTT have theoretical promise their effectiveness is unclear These technologies also raise important ethical legal and governance challenges that require comprehensive analysis in order to support decision-making regarding their appropriate use A number of frameworks recommendations and analyses have emerged recently in an effort to chart potentially ldquosaferdquo pathways for use of public health disease surveillance technology Many in the United States such as the Electronic Frontier Foundation Elec-tronic Privacy Information Center American Civil Liberties Union and the Center for American Progress are proposing that digital public health surveillance technologies must embrace strict data privacy protections decentralized data storage a high degree of anonymity and voluntary adoption (Crocker Opsahl and Cyphers 2020 Electronic Privacy Infor-mation Center 2020 Kahn Gilmor 2020 Simpson and Conner 2020) Others have argued that technologies that seek to enhance public health response during a pandemic should more closely align with the needs of public health professionals and the evidence-based procedures they follow stating that interests in serving the publicrsquos health ought to weigh more heavily in the necessary balancing of stakeholder interests (de Jong et al 2019 Watson et al 2020) This view is in part based on a recogni-tion that during countless other outbreaks the public has benefited from traditional disease surveillance and contact tracing which are heavily re-liant on centralized data storage and when necessary the collection of identifiable information These traditional approaches are governed by ethics principles (PHLS 2002) ethics guidelines (WHO 2017) and laws (ASTHO 2012) and digital technologies represent a new tool to support them
While debates and recommendations about appropriate design and use of DCTT have focused intensely on minimizing important data-related risks a wider lens is needed to fully appreciate the many additional criti-cal questions that need attention This report begins to grapple with these questions which are critical to address in order to guide responsible use of DCTT Given the complexity of the terrain as a first step toward estab-lishing a foundation for responsible decision-making regarding potential use of DCTT we offer a set of guiding principles (see box) These prin-ciples are meant to apply to DCTT as well as other digital technologies used in novel ways during pandemic response
Guiding Principles for the Use of Digital Public Health Technologies for Pandemic Response
Transparency and public engagement are essential to an inclusive digital public health response
bull Government public health and digital technology leaders must engage
effectively with the public and other stakeholders to communicate the util-
ity importance oversight and limitations of relevant digital technologies
including their implications for individualsrsquo privacy and civil liberties
bull Transparency at all levels is essential for maintaining public trust and
confidence
bull To the extent possible digital public health responses should reflect the
range of values that are important to individuals including advancing the
health and well-being of the community as a whole
bull Decision makers should recognize the sacrifices that some people may be
willing to make during a pandemic in order to advance public health goals
Acceptance by some of particular monitoring capabilities should not be read
as a willingness to extend these methods to other problems or uses
Digital public health responses must represent the least infringement of civil liberties necessary to accomplish the public health goals
bull If preferred digital public health strategies infringe on privacy and other civil
liberties the infringements must be sufficiently justified by the circum-
stances of the pandemic offset by ample anticipated public benefit and
considered relative to infringements associated with other possible strate-
gies such as mass physical distancing
bull Only those data that are necessary and relevant for the stated public health
purposes should be collected Identifiable data should be stored in a secure
manner and only for the period of time that the public health purposes
require
bull Adopted technologies should not be used in ways that subject communities
to discrimination or surveillance for nonndashpublic health reasons
bull Respect for individual autonomy requires that users are sufficiently informed
of the public health goals of the technology and the extent to which those
goals are being met
Use of digital public health technologies and data must be guided by best available evidence
bull Decisions to deploy digital public health technologies should be based on a
careful assessment of the uses and limitations of any proposed technology
taking into account the best available evidence
bull Those who deploy digital public health technologies should continuously and
systematically monitor their performance as well as any evidence that is being
generated in other contexts about the selected technological solution and about
other competing technologies
bull Unintended consequencesmdashincluding those that might impact public health
goals core values and interests of the public and unfair advantage or disadvan-
tagemdashshould be carefully monitored and addressed as necessary
Responsible use of digital public health technology requires meaningful governance and accountability
bull Systems of governance must be trustworthy and well informed They must be
reviewed and adjusted as circumstances and evidence change or as unintended
effects are identified
bull Trusted representatives who are capable of developing and implementing uni-
form and fair standards for adopting and utilizing underlying digital technology
must be identified
bull Understandable transparent and publicly accessible rules must guide the
collection access control use storage and combination of data by government
authorities public and private institutions and other parties such as public
health researchers
bull Oversight accountability and consequences for abuse or misuse of these data
must be explicit and enforceable
The deployment of digital public health technology must be rooted in a commitment to equity
bull Digital public health technologies should be deployed in a manner that does
not propagate preexisting patterns of unfair disadvantage or further distribute
harms and risks unfairly throughout the population
bull To the extent possible digital public health technologies should be designed to
rectify existing inequities
bull Oversight mechanisms must be in place to ensure that the improved public
health outcomes are equitable and to detect and correct any unforeseen resul-
tant injustices attributable to the technology or that can be addressed using the
technology
bull The incentives and disincentives for adopting new technology must be equita-
ble not exploitative and aligned with effective use of the technology
bull Disparity-driven technology gaps should be explicitly recognized To the extent
possible provisions should be made to address the digital divide
Introduction 27
In reflecting on these principles it becomes clear that if we wish not only to realize but to maximize the public good that might come from use of DCTT we must carefully define and responsibly align public health needs and capabilities with technological needs and capabilities We must understand that although technology may serve as a workforce multiplier it alone will not solve the public health challenges we face We must identify and address assumptions and misinformation about technologies and data use We must provide the means and opportu-nity for informed decision-making by the public and those who serve as our representatives Government officials public health leaders leaders of other institutions employers digital technology developers and the public all must be adequately informed and engaged in order to make the best decisions possible under the circumstances
ONE
29
Types of Information Collected through Contact Tracing
Data Collected from Infected Persons
Symptoms and Course of Illness
Information about COVID-19 patientsrsquo signs symptoms and course of illness is important to public health because it provides a basis for refining clinical case definitions and informing health care providers and the gen-eral public (CDC 2020c) This includes the specific signs and symptoms manifested by persons who are COV+ as well as the relative frequency and durations of different signs and symptoms This would also take into consideration those persons with no symptoms but who test positivemdashthose who are presymptomatic (develop symptoms later) those who are postsymptomatic (clinically recovered but still infectious) and those who never manifest illness at all
Typically contact tracing begins with a case in which a person has confirmation of infection by means of a diagnostic test However in some cases test results are not reported until several days later and individuals may be identified as ldquopresumptive positiverdquo cases until testing can be completed In these cases contact tracing efforts will need to be updated when test results are returned For example if a test comes back negative public health professionals will want to notify contacts that they no lon-ger need to quarantine
Public Health Perspective
30 Digital Contact Tracing for Pandemic Response
Movement and Contacts
In order to manage cases appropriately (identify and track the infected isolate the sick quarantine the exposed) public health officials need de-tails on each case (Resolve to Save Lives nd) First they need to know who and where the individual is That means personally identifiable infor-mation and contact information (address phone numbers email) It also means information about the nature intensity and duration of contact with individuals to whom they may have transmitted the disease This may include information about where the individual works and the kind of work they do (eg health care worker) how they travel (eg bus sub-way car) and where they shop or any other public venues they may have visited during a period of possible infectiousness (PIH 2020a) It may be helpful in certain circumstances for public health officials to ensure that suspected cases contacts or other high-risk individuals are following iso-lation and quarantine recommendations or orders
Contact tracing involves identifying all individuals who have had sig-nificant exposure to confirmed or probable cases during the time prior to and after the onset of symptoms both of which are times when the case is thought to be infectious (Africa CDC 2020) Contacts could be those who are caring for COVID-19 patients especially if they lacked proper PPE and those who had close interaction with the COV+ person over a sustained period of time particularly in enclosed spaces (PIH 2020a) For COVID-19 contacts are identified by asking a person with a confirmed or probable case about people they may have been within 6 feet of for 15 minutes or more starting from 48 hours before the onset of symptoms and lasting until the person is isolated (CDC 2020b)
Data Collected from Contacts of Infected Persons
Contact Details
In addition to the data collected from individuals with COVID-19 con-tact tracers will collect data from potentially exposed individuals (con-tacts) Information about the nature intensity and duration of contact with an infected person may be collected for a contact if information about the case is known to the contact These details can help a contact tracer more accurately determine whether the contact is at high or low risk for SARS-CoV-2 transmission and help determine whether a con-tact should quarantine for 14 days (the upper bound of the SARS-CoV-2
Public Health Perspective 31
incubation period) In addition public health professionals may gather contactsrsquo demographic information and other personal data to contribute to population-level disease surveillance and situational awareness about an epidemic (CDC 2005) However the information needed at baseline is only a personrsquos name and contact information
Symptoms (If They Develop) and Course of Illness as well as Information about Close Contacts
If a contact develops COVID-19 symptoms while in quarantine andor tests positive for the virus public health will then collect the data required for a COVID-19 case This includes collecting information on the con-tacts that a person may have had (if any) in the days immediately before and during the course of their infection
How Contact Tracing Information Informs Public Health Action
To reduce disease burden and help make ldquoreopeningrdquo safer during the COVID-19 pandemic the United States and other countries will need to identify gather information about and safely isolate cases and quar-antine their contacts to reduce community transmission (Watson et al 2020) Gathering information about possible cases and their contacts en-ables public health to break chains of transmission
Contact tracing involves stages (CDC 2020a) including
1 identifying an infected person as a COVID-19 case
2 identifying the close contacts of that case (Africa CDC 2020)
3 getting in touch with contacts
4 asking contacts to quarantine at home for 14 days
5 assessing contacts for possible symptoms and
6 following up with COV+ persons and their contacts to identify new or worsening symptoms and connect them with medical care if needed
Contact tracers also play an important role in providing resources for COV+ persons who are in home isolation and their contacts who are
32 Digital Contact Tracing for Pandemic Response
in home quarantine Knowing who and where cases and contacts are can enable provision of supplies such as digital thermometers or masks Effective contact tracing that enables isolated cases and quarantined con-tacts to remain at home also requires providing a range of social sup-port services or ldquocare packagesrdquo from delivering food and medicines to trash pickup Furthermore vulnerable individuals who are homeless or otherwise unable to sufficiently isolate or quarantine in their current living conditions may need to have alternative housing arranged to safely remain separated from others (CDC 2020b)
Finally contact tracers explain what quarantined contacts should do if they begin to develop symptoms consistent with COVID-19 (Africa CDC 2020) Depending on the context contact tracers may engage in active monitoring by regularly communicating with contacts about their health status through phone text message or possibly mobile applica-tions In rare cases public health can make quarantine mandatory and may monitor a quarantined individual to ensure that they do not break quarantine Contact tracers may also facilitate access to health care by providing telemedicine resources or other information and support for accessing medical care
Characteristics That Make Data Useful to Public Health for Reducing Disease Transmission
Data Access
If digital contact tracing technology and closely related digital health products (together DCTT) are intended to support the public health ac-tions described above and directly amplify public health capacity to con-duct case identification and contact tracing then data collected through DCTT must be accessible to public health authorities Identifying infor-mation and location data for cases and contacts of cases are necessary for public health use so that contact tracers can do their work to uncover ongoing transmission and enable isolation and quarantine These data should also be durable meaning that public health can return to the data in order to interact with and support cases and contacts These data can also be useful at a population level if de-identified and aggregated by illuminating trends in community transmission and providing support for decisions about resource allocation
Public Health Perspective 33
Data Format
Data should be provided to public health authorities in a usable format that is compatible with public health systems and that has the granularity and specificity of personal information that is needed for use in contact tracing Without personal identifiers the data cannot be used by public health workers to undertake contact tracing Data should also contain information about the nature of a contact including the proximity of the contact and number of minutes that the person was in contact with an infected individual Location data can also help public health author-ities to conduct contact tracing particularly when contact occurred in a crowded area and involved people who donrsquot know one another Loca-tion data from a case can help public health professionals identify con-tacts even when those contacts themselves are not using a contact tracing app because the data shows contact tracers where to look for additional contacts
Data Accuracy
Data that identifies individuals as having sustained contact with a case must be as accurate as possible If criteria for being considered a contact are too restrictive it may result in missed contacts and sustained chains of disease transmission If criteria are too broad it may result in unneces-sary restriction of movement which could have significant personal and economic consequences
Timeliness of Data
Data from cases and contacts must be timely in order to enable case-based management that will help reduce community transmission For contact tracing to be effective infected individuals need to be isolated and their contacts identified and quarantined as quickly as possible Testing for SARS-CoV-2 can take time sometimes many days for a test result Especially because SARS-CoV-2 is transmissible during the pre-symptomatic period data on symptomatic individuals should be made available to public health officials even before a positive test is returned in order to enable identification and quarantine of contacts right away If this information is delayed until a test result is received it may be too late to identify and quarantine contacts because contacts (if infected) will already be contagious and may have spread the virus to others
34 Digital Contact Tracing for Pandemic Response
VolumeAvailability of Data
The more that individuals opt to share their information to support con-tact tracing the more effective contact tracing will be in breaking chains of viral transmission and controlling epidemics of COVID-19 The exact proportion of cases and contacts that need to be identified in order to avoid large surges of cases which overwhelm health care systems is un-certain but the goal is to identify all infected cases and all close contacts of each case (PIH 2020b)
Recommendations
bull Technologies or apps with the goal of enhancing public health ca-pacity to identify cases and trace contacts in order to control the spread of SARS-CoV-2 should be designed to match functionality with that goal
bull Technologies or apps may produce some false negatives or false positives but they should be accurate enough that public health authorities feel confident that they support and donrsquot detract from contact tracing efforts
bull DCTT approaches for public health should be designed to facilitate the following
deg identifying contacts including those who may not be easily found otherwise
deg finding and notifying contacts rapidly before they develop symptoms if infected
deg analyzing the nature of contact to determine whether contact is high medium or low risk and to support decisions about whether a contact should quarantine and
deg following up with cases and contacts so that public health can provide resources to support isolation and quarantine
bull Data collected through DCTT should be made available to public health professionals and to researchers in de-identified form to sup-port population-level epidemiologic analysis
T WO
35
Characteristics of SARS-CoV-2 Relevant to Candidate Digital Solutions
The SARS-CoV-2 virus has some unique transmission characteristics and clinical manifestations that can help guide use of digital contact trac-ing solutions Individuals infected with this virus may or may not show symptoms or may show a range of different and sometimes nonspecific symptoms Estimates regarding the percentage of individuals who are in-fected but never develop symptoms is highly uncertain ranging from 5 to 50 (Heneghan Brassey and Jefferson 2020) Data suggest that a substantial proportion of transmissionsmdashperhaps as high as 50mdashoccur between individuals who are not symptomatic and that transmissibility may extend out as long as 3 days before the onset of symptoms (WHO 2020)
The complexity of asymptomatic and presymptomatic transmission makes it more difficult to identify all cases of COVID-19 It also means that manual contact tracing is less effective because people are unlikely to remember all of their contacts during the long period of infectivity (Ferretti et al 2020) however it does not negate the need for contact tracing Identifying symptomatic cases will still help greatly with slowing the spread because their contacts can be asked to quarantine to prevent them from spreading the virus if they are indeed infected This means that whether they become symptomatic or not contacts will be quarantined and the chain of transmission will be broken If contact tracing can be implemented on a large enough scale perhaps with support from DCTT
Digital Technology and Contact Tracing
36 Digital Contact Tracing for Pandemic Response
eventually the virus could be managed at much lower levels of community transmission and large epidemics of unrecognized spread will not occur
The transmissibility of the virus when a person has no symptoms further suggests that effective solutions may require multimodal inter-ventions combining contact tracing with frequent rapid and ubiquitous testing and continued social distancing to varying extents (Cheng et al 2020)
Because of presymptomatic spread contact tracing efforts and dig-ital solutions to augment those efforts should support identification of contacts a person had 2 days before their symptoms and at least 3 days after the resolution of those symptoms (if the person continued to have contacts through that time period) (CDC 2020d) Additionally public health messages delivered by these technologies should urge contacts to quarantine for the full 14-day incubation period
Previously Existing Contact Tracing Technologies
Prior to this pandemic health agencies in high- medium- and low-income countries had begun to develop and use digital tools to augment the man-agement of infectious diseases including sexually transmitted infections (HIV chlamydia gonorrhea) and high-consequence epidemics (Ebola) (Danquah et al 2019)) However these have been primarily used to facil-itate case interviews partner notification (in the case of STIs) and record keeping as opposed to fully digitizing or automating the contact tracing process
It has been recently suggested that digital contact tracing could con-tribute to the management of the ongoing COVID-19 pandemic and the experiences of containing SARS-CoV-2 in countries such as China Sin-gapore and South Korea provide noteworthy examples However un-dertaking this case-based intervention on the scale required to achieve pandemic control is a novelty in the history of public health Although technological development is proceeding rapidly several foundational is-sues have yet to be resolved including functionality connectivity to pub-lic health authorities and informatics systems usability by disease inter-vention specialists (DIS also referred to as contact tracers) and sufficient protection of personally identifiable information among others
Digital Technology and Contact Tracing 37
Introduction of Novel Digital Contact Tracing Technologies
Digital contact tracing technologies and platforms have recently been in-troduced and the CDC has published preliminary criteria for evaluating these tools (CDC 2020e) It can be helpful to consider three broad ap-proaches along the spectrum of potential methods of digital contact trac-ing a maximal approach (typified by the South Korean governmentrsquos cen-tralized data collection (M S Kim 2020)) a minimal approach (typified by the AppleGoogle decentralized privacy-protecting proximity tracking (Apple and Google nd)) and a diverse middle ground that aims to aug-ment manual contact tracing with the collection of digital data Perhaps the most promising approach in this middle ground involves allowing us-ers to turn over both proximity data and GPS location data (ie cell-site location data) to public health authorities on a voluntary basis
Along with this ldquominimal to maximalrdquo spectrum in the design of dig-ital contact tracing technologies and systems there is another spectrum that concerns voluntary versus mandatory use of these technologies are individuals entirely free to use these technologies or not or should poli-cies incentivize or even mandate their use At one extreme South Korea (Republic of Korea) implemented a system (called Safe Korea) supported by the Ministry of the Interior and Safety that collects a variety of per-sonal data in a centralized database in order to enforce quarantine orders and track possible contacts (M S Kim 2020) Israel also implemented a centralized involuntary data collection system for tracking COVID-19 cases and alerting those who may have been exposed (Hendrix and Eg-lash 2020) In Poland health authorities have set up mandatory ldquocheck-insrdquo involving a GPS-waypoint capture and ldquoselfierdquo photographs sent to the monitoring agency to ensure that individuals are not breaking quar-antine (Hamilton 2020)
These centralized systems can be designed to incorporate data from a variety of sources The data collected include location data from mobile phones QR codes can also be scanned to track the use of public transit where GPS data may be inadequate (due to low resolution) to accurately distinguish the occupants of one vehicle from another The data collected from mobile phones can then be integrated with data from other sources such as facial-recognition cameras credit card transactions and social media
38 Digital Contact Tracing for Pandemic Response
At the other extreme of technology invasiveness for contact trac-ing isolation and quarantine many corporations and working groups (including the AppleGoogle collaboration) have developed privacy-pre-serving proximity tracking (PPPT) using Bluetooth Low Energy (BLE) ldquohandshakesrdquo that record close contact between mobile phone users In most architectures these proximity data are stored in the usersrsquo phones as anonymized ldquobeaconsrdquo that cannot be used to re-identify the users directly If a user with a PPPT app installed on their phone tests positive and enters test results into their app those who have been identified as having been in close proximity to them can be notified by the app This notification can be automatic or at the discretion of the person who is COV+ depending on the app design If notified a user who has been in contact with a COV+ individual would receive a push notification alert-ing them to possible exposure (which may be timestamped) but with no other identifying information
Because of its reliance on anonymized data PPPT on its own is dis-tinct from manual contact tracing In recognition of this fact some de-signers and researchers now use the more descriptive term ldquoexposure no-tificationrdquo Moreover the public health usefulness of PPPT is uncertain it is unclear how PPPT can best be used in tandem with manual contact tracing especially if the data it collects are inaccessible to or unusable by public health authorities It remains to be seen whether PPPT will provide significant benefit operating alongside but not integrated into manual contact tracing
Between these extremes there are a number of possible middle-ground approaches that aim to strike a balance among public health utility tech-nological feasibility and user privacy protections This middle ground divides into two rough categories centralized storage of de-identified data and decentralized storage of personally identifying data The United Kingdomrsquos NHSX is reportedly developing an app that would utilize BLE handshakes to collect anonymized proximity data which would then be stored on a centralized government-operated server
The most prevalent middle-ground approach in the United States context involves the collection and storage of personal datamdashincluding identifying information and location datamdashon the userrsquos phone This de-centralized but personally identifiable data can then be voluntarily shared with public health officials if the user tests positive for SARS-CoV-2 For
Digital Technology and Contact Tracing 39
example an MIT team has developed an app called Private Kit Safe Paths (MIT nd) that stores usersrsquo location data on their phone for 28 days If a user tests positive she can voluntarily upload her location data to a website that is accessible only to public health officials Officials can then analyze this personally identifiable data and subsequently broadcast re-dacted and de-identified data to other users Healthy users would have access to these redacted location data of COV+ users but their own data would not leave their phones (The developers plan to incorporate BLE proximity data once available) Along similar lines the North Dakota state government has rolled out an app that stores both location data and proximity data on a userrsquos phone which can be voluntarily released by the user to public health authorities if the user tests positive (NDDoH 2020) At a minimum the storage of user location data can function as a ldquomemory aidrdquo if the user tests positive But releasing the data to public health authorities may help them analyze the spread of COVID-19 and alert individuals or groups that have been in contact with persons who are COV+ An overview of various DCTT apps and platforms as well as features that are relevant to this analysis are provided in Table 1
Because DCTTs are so new very little is known about their actual utility to public health authorities for controlling this pandemic Although multiple countries that have had success in greatly reducing transmission of SARS-CoV-2 have included DCTT in their response these countries have employed multiple simultaneous approaches to controlling the vi-rus including manual contact tracing and it is difficult to disentangle what made those responses successful Preliminary impressions from Ice-land may suggest that DCTT at least in that context had a small impact on reducing transmission ldquoespecially compared with methods of manual contact tracing such as phone callsrdquo (Hadavas 2020) This is with the highest public download rate of any DCTT app thus far
DCTTs have the potential to be helpful but they also have the poten-tial to distract from other public health efforts including manual contact tracing Concerns about implementation of DCTT from the public health perspective include that data generated may not be useful to public health authorities either because they donrsquot include detailed data to aid contact tracers or because the data are inaccurate (Mills Rodrigo 2020) DCTT if not calibrated well could be overly inclusive and create many false positives This would be harmful to those individuals being notified and
Purp
ose
Tech
nolo
gies
Use
dD
ata
Stor
age
Part
icip
atio
n
Proximity-based exposure notification
Digital contact tracing (DCT)
Bluetooth LE
GPS
SMS
Centralized
Decentralized
Mandator (actually or functionally)
Voluntaryopt-in
Gov
ernm
ent a
cces
sA
pp N
ame
Dev
elop
er o
r Cou
ntry
Max
WeC
hat
Alip
ayCh
ina
D
ata
com
es fr
om g
over
nmen
t so
urce
s lo
catio
n da
ta s
ent t
o po
lice
Intervention Type
Middle Ground
Trac
e To
geth
erSi
ngap
ore
M
anda
tory
gov
ernm
ent
acce
ss if
pos
itive
NH
SXO
xfor
dO
xfor
d
Gov
ernm
ent m
aint
ains
dat
a
Nex
tTra
ceFr
ed H
utch
inso
n Ca
ncer
Re
sear
ch C
ente
r
Gov
ernm
ent m
aint
ains
dat
a
but n
o st
orag
e
COVI
D
Safe
Path
sM
IT
Volu
ntar
y up
load
by
user
s
who
test
pos
itive
Aar
ogya
Set
uIn
dia
A
nony
miz
ed a
ggre
gate
Care
19N
orth
Dak
ota
In
agg
rega
te o
ptio
nal i
f pos
itive
Minimal
Covi
dSaf
eU
niv
of W
ashi
ngto
n
Non
e
Covi
dWat
chU
niv
of S
tanf
ord
amp U
niv
of W
ater
loo
To
val
idat
e te
st re
sults
CoEp
iCo
Epi
O
pt-in
to s
hare
BT
and
sym
ptom
lo
g w
ith C
oEpi
ser
ver
itoG
erm
any
N
one
pos
itive
resu
lts to
ito
serv
er
TAB
LE 1
Ex
ampl
es o
f Dig
ital C
onta
ct T
raci
ng T
echn
olog
ies
to S
uppo
rt A
ctiv
e Pu
blic
Hea
lth S
urve
illan
ce a
nd R
elev
ant F
eatu
res
Digital Technology and Contact Tracing 41
asked to quarantine unnecessarily and it could result in large proportions of the population remaining at home at any one time Individuals living or working in congregate settings could receive frequent notifications that would result in their inability to leave quarantine for long periods of time Finally public health authorities could also become inundated by data from these technologies and not have sufficient approaches to manage or analyze the incoming information
Relevant Differences between Manual and Digital Contact Tracing
There are several noteworthy differences between manual contact tracing efforts and use of DCTT First there is a significant amount of evidence regarding the effectiveness of manual contact tracing which is lacking for DCTT Second manual contact tracers interact with individuals who are confirmed or suspected cases and contacts of cases but not other members of the general public DCTT intervention would affect all users regardless of circumstances (though some more than others) Third manual contact tracing occurs most often through human-to-human encounters with the opportunity to clarify misconceptions address worries and express sympathy and other important affects DCTT can certainly incorporate sharing of important information and potentially communicate some af-fect but it currently lacks a range of other human capabilities and char-acteristics Fourth there typically are fewer data intermediaries in manual contact tracing (fewer entities handling data) in DCTT a valid argument could be made that a wide range of technology developers (and perhaps mobile network operators) must remain connected to relevant data in order to continuously identify problems and improve functionality
It is because of these and other differences that DCTT has been pro-posed as a potential complement to rather than a replacement for man-ual contact tracing However over time it is possible that technology could develop to close gaps between some of these differences (if and as needed) and in parallel the goals of contact tracing and public health surveillance may evolve
Ethics of Designing and Using DCTT
43
THREE
43
Those developing DCTT and those considering its use should systemat-ically take into account and document alignment with the guiding princi-ples outlined in this report
When considering the ethics of DCTT key ethical questions con-cern the features that DCTT should have (eg should digital contact tracing apps collect usersrsquo location data) whether and how individualsrsquo data should be shared with public health authorities how ethically to encourage use of DCTT (eg under what circumstances would it be eth-ical to incentivize or mandate use of DCTT) what kind of supports and equity-promoting measures should accompany use of DCTT and how governance and oversight of DCTT should be structured
The sections that follow consider these questions one by one A key conclusion of this report is that these features of the design and use of DCTT are ethically interrelatedmdashreaching a determination regarding any one question requires careful consideration of them all Rather than reaching ldquoone size fits allrdquo conclusions about specific features of uses of DCTT decision makers should ethically assess DCTT systems holistically
Generally a public health measure is ethically justifiable if it strikes a reasonable balance between competing considerations and if it pro-vides sufficient public health benefit (or the prospect of benefit) to justify the burdens associated with it DCTT systems are ethically justifiable if they strike a reasonable balance between multiple ethical considerations including
bull enabling an effective and efficient public health response
bull protecting individual privacy and preventing harms to individuals
44 Digital Contact Tracing for Pandemic Response
including harms from sensitive data being revealed and from erro-neously being subjected to isolation or quarantine orders
bull allowing individuals to control what information about them is collected and revealed to whom including through appropriate dis-closure and authorization processes for data collection
bull promoting equitable distribution of benefits and burdens of DCTT
bull maintaining public trust in DCTT and in the COVID-19 public health response and
bull taking seriously the future implications of decisions that we make today
To illustrate a holistic assessment consider whether it is ethically jus-tifiable for an employer to mandate that employees use a DCTT as a condition of returning to work This will depend upon many features of the DCTT system what kind of data the DCTT collects (eg does it collect location data or just record proximity events) whether there is public health capacity to make good use of these data what the data are used for (eg will the employer ban an employee from the workplace on the basis of a DCTT-identified contact) what kind of social supports are available (eg is there paid leave for employees) what employeesrsquo attitudes are toward use of DCTT and whether mandating use is likely to have public health benefit among other factors These factors may vary from place to place and may change over the course of the pandemic Thus there is no ldquoone size fits allrdquo ethically optimal approach to DCTT
Justifying the Use of DCTT Systems
A foundational issue is why deploying any DCTT during a pandemic is justified given there are manual contact tracing capabilities that are well established while the performance and effectiveness of novel technolo-gies is less established The need to move quickly to minimize the spread of the virus poses challenges here as the data needed to fully make the case that these technologies substantially contribute to the public health response may not be available prior to widespread use The primary ar-gument for DCTT is that the capacity of manual contact tracing may be
Ethics of Designing and Using DCTT 45
exceeded and we may not be able to bolster the public health workforce rapidly and sufficiently enough to meet needs DCTT has the potential to quickly and exponentially expand the reach of contact tracing In ad-dition DCTT may allow more efficient identification and quarantine of potential contacts of COV+ people than manual contact tracing alone particularly given the high number of infections that have been spread by asymptomatic individuals
Nonetheless reasonable people disagree about the prudence of pur-suing DCTT especially given its limited performance history and poten-tial risks including diverting attention and resources from more effective interventions The limited attention and resources available during a pan-demic must be allocated efficiently and effectively
To justify potentially widespread use of technologies such as DCTT therefore a number of considerations must be addressed
bull whether the technology is designed to meet an important and unmet public health need
bull whether there is sufficient evidence or reason to suggest that the technology will be effective at serving its purpose
bull whether the outbreak is characterized by sufficiently severe morbid-ity and mortality and a high rate of disease transmission to warrant large-scale introduction of novel systems
bull whether there are other less autonomy-restricting or less risky al-ternatives to widespread use and
bull whether it is reasonably likely that a sufficient number of individ-uals will use the technology to achieve the intended public health benefit
Monitoring and Evaluating Technologies to Inform Policy and Practice
A number of public health ethics principles necessitate the ongoing mon-itoring and evaluation of DCTT systems First DCTT must be shown to perform reasonably well at achieving its stated goal reducing the spread of SARS-CoV-2 The effectiveness of DCTT programs should be illus-trated at a number of stages
46 Digital Contact Tracing for Pandemic Response
1 Robust initial technology testing is needed to publicly justify the widespread adoption of DCTT and avoid public failures which may hamper future uptake (eg Lovejoy 2020 Morse 2020) This typically includes alpha testing in virtual environments and beta testing in different community settings
2 If and when a DCTT is implemented on a wide scale it must be monitored on an ongoing basis to assess reach effectiveness func-tionality best practices and any harms
3 When approaching a previously identified stopping point for use of DCTT monitoring can help to identify when utilization is no longer needed
If at any of these points evidence clearly suggests harm (particularly in comparison to other methods that the public might find more acceptable) this evidence should provide a basis upon which to revisit strategies pri-orities and allocation of resources Attention should be given to foresee-able side effects that may dramatically influence the overall effectiveness of the program such as individuals carrying their smartphones around with them selectively so as to avoid particular undesired consequences of DCTT policies
Anonymized aggregate data including user feedback must be eval-uated to ensure that benefits and burdens are distributed fairly As noted earlier unintended burdens may include inequitable outcomes that may arise in a DCTT program for example resulting from uneven access to the required technology to participate disparate concerns about sur-veillance within some communities that might limit widespread use or discrimination that may result from being identified as COV+ due to the program or for communities that are termed ldquohotspotsrdquo based on maps of COV+ location data Additionally it is possible that some communi-ties might get higher rates of false positives because they are located in densely populated areas thus increasing the burden of self-isolation If any of these inequities are identified steps must be taken to mitigate them
Finally numerous actors should engage in the monitoring and eval-uation of DCTT systems Technology developers and public health re-searchers have a clear role in this process Technology developers should work with public health researchers to monitor accuracy precision func-
Ethics of Designing and Using DCTT 47
tionality confidence of estimates sources of error and the like Research-ers may also be able to contribute innovative methods to systematically and rapidly evaluate candidate technologies such as by deploying cluster randomized stepped wedge (Hemming et al 2015) or adaptive trial de-signs and techniques (eg response-adjusted randomization) (Pallmann et al 2018) These approaches were also proposed for use in research to assign candidate experimental treatments and vaccines during the 2014-15 Ebola outbreak (Berry et al 2016) When formal research activities are pursued ethics principles and legal requirements for the conduct of research should apply (eg The Belmont Report)
Furthermore any workplace or institution that incentivizes or man-dates use of DCTT has a responsibility to provide evidence that the in-tervention at minimum is not likely to cause harm and to monitor for unanticipated burdens In all cases it is vital that a trusted intermediary be involved in the evaluation of DCTT programs to limit perceptions of bias and ensure a legitimate basis for decision-making Nonsensitive aggregate DCTT analyses should be made available to the public so as to permit verification and inform continuing public debates about its useful-ness and necessity At an individual level data should be available to users that would permit them to further investigate their personal risk with public health officials or other health workers This is important not only to ensure their health and well-being but also to add a layer of protection against unnecessary quarantine
Recommendations
bull Reviews of DCTT systems must be conducted in part by an inde-pendent intermediary that has established the publicrsquos trust
bull Those who authorize use of DCTT within a particular jurisdic-tion or institution should continuously and systematically monitor the technologyrsquos performance in that context This should include monitoring for effectiveness and benefit monitoring for harms and monitoring for the fair distribution of both benefits and harms They should also monitor evidence that is being generated in other contexts about their selected technological solution and about other competing technologies
48 Digital Contact Tracing for Pandemic Response
bull Data should be available to users that would permit them to further investigate their personal risk with public health officials or other health workers to add a layer of protection against unnecessary quarantine
Public Trust and Public Attitudes
Researchers have estimated perhaps conservatively that DCTT use by 80 of smartphone ownersmdash56 of the populationmdashwill be needed to suppress the epidemic (Hinch et al 2020) These estimates also highlight that some decrease in transmission would be realizable even with lower rates of technology adoption As such in order to maximize impact it is essential to gain a thorough understanding of public perspectives on DCTT including which features and uses of the technology the pub-lic finds acceptable which kinds of DCTT the public would be most likely to use and which designs and uses of DCTT would maintain or jeopardize public confidence and trust There will be variation in public attitudes within and across societies and over time
With respect to what we currently know about public attitudes and trust in DCTT in the United States polling data suggest some potential support and also some divisions regarding willingness to use the technol-ogy Polls conducted by groups based at the University of Zurich (Hargit-tai et al 2020) and the University of Oxford (Altmann et al 2020) suggest that more than 60 of Americans would be willing to install such an app Both a Washington PostndashUniversity of Maryland poll (2020) and a Kaiser Family Foundation poll (Kirzinger et al 2020) show roughly half of the population would be willing to install the app Over half of the population (59) would be willing to share their COVID-19 positive test result with an app in order to anonymously share that information with their contacts (Washington PostndashUMD 2020) Only 29 of respondents to a March 12ndash27 Oliver Wyman Forum poll (Elliott et al 2020) said that they would be willing to share their location data Additionally Washing-ton PostndashUMD data and Pew data from 2019 suggest that approximately one in six Americans do not have a smartphone and thus cannot use the technology without intervention (Pew Research Center 2020)
People may be more willing however to download an app if it will
Ethics of Designing and Using DCTT 49
ease social distancing policies and allow for more economic and social activity Willingness to install a contact tracing app increased among re-spondents to the Kaiser Family Foundation poll from 50 to 66 when respondents were asked if they would be willing to do so to allow schools and businesses to reopen Additionally who develops or administers the app appears to matter Respondents to the Washington PostndashUMD poll indicated higher levels of trust that their anonymity would be preserved by public health agencies and universities than by tech companies or health insurance companies Further more respondents to the Oliver Wyman Forum poll were willing to share their health information with public health authorities (55) than the local government (35) their employer or school (33) or the federal government (27)
These data suggest that people will be more willing to use a contact tracing app when the potential benefits are clearly identified and valued such as lifting social distancing measures and they will be more willing to do so if the data are going to a public health agency rather than the federal government or a tech company Other factors that seem to be as-sociated with greater willingness to install a contact tracing app include younger age and the app source (Hargittai et al 2020) with a preference for apps distributed by public health agencies over others such as health insurers or public universities (Hargittai and Redmiles 2020) However all of this must be read with caution as public polling may not be repre-sentative of some populations or of widespread public attitudes Further these attitudes may shift over time and may be discordant with behaviors (Barth and de Jong 2017)
Deliberative public engagement efforts would be an appropriate means of filling in gaps in understanding about the acceptability of dif-ferent approaches (Fishkin and Laslett 2003 Cavalier 2011) In addi-tion including the public particularly in the earlier stages of planning a path to sustainable resolution to the pandemic could serve to help disseminate a nuanced understanding of what is at stake including the key challenges and trade-offs Aggregated public polling results are not sufficient as a proxy for careful analyses of the ethical challenges but they do provide a necessary input for these analyses Integrating lessons and outputs from public engagement into guidance and other products requires special attention and should be validated and enhanced through further engagement
50 Digital Contact Tracing for Pandemic Response
Recommendations
bull More research into public attitudes is needed In particular in-depth qualitative research should examine public attitudes about percep-tions of trust in DCTT among different communities which fea-tures of DCTT influence trust and the extent to which people are willing to provide different types of data through DCTT to help their community
bull States and localities that are considering adopting DCTT should engage with the public to increase their understanding of the ac-ceptability of DCTT design features and uses among diverse communities
Designing Flexible Technology to Maximize Public Health Utility While Respecting Other Values
Values in Design
Efforts to advance DCTT in the United States and elsewhere have empha-sized the importance of ldquoprivacy by designrdquo that is building privacy and security protections into the design of technology rather than counting on responsible use alone (Cavoukian 2010) As noted above some major technology companies have signaled this position through development of decentralized privacy-preserving proximity tracking (PPPT) systems These systems embed features such as decentralization anonymity of us-ers bans on collection of location data and minimal reliance on or inte-gration of public health authorities or other government actors Many of these features have also been embraced early by advocacy organizations (Crocker Opsahl and Cyphers 2020 Electronic Privacy Information Center 2020 Kahn Gilmor 2020) and in an open letter from nearly 300 researchers (ldquoJoint Statement on Contact Tracingrdquo 2020)
Privacy by design provides principles that incorporate one set of val-ues (privacy) into the design of DCTT Importantly the principles ac-knowledge the need to design privacy defaults into systems while main-taining the capacity of those systems to achieve their otherwise justifiable ends Put another way privacy by design ldquoembraces legitimate non-pri-vacy objectives and accommodates them in an innovative positive-sum mannerrdquo (Cavoukian 2010 p 4)
Ethics of Designing and Using DCTT 51
This stance simple in its statement is not easy to satisfy Given that ldquoobjectivesrdquo are themselves driven by values it begs for an articulation of additional values (aside from privacy) that individuals and groups within societymdashincluding many privacy advocatesmdashmay believe to be important For example at any moment in addition to valuing their own privacy individuals may value efficiency equity autonomy economic well-being companionship patriotism or solidarity Moreover the above stance necessitates an acknowledgment that peoplesrsquo value priorities often change when circumstances change not least of which during a pandemic when mass physical distancing has made it difficult to fully realize many important values (aside from physical privacy) A different orientation is needed at this moment As Flanagan Howe and Nissenbaum (2008) conceptualized in 2008 we should take a ldquovalues in designrdquo approach to DCTTmdashan approach that designs a broader range of values such as those enumerated above into technology
This approach requires a wider ethical lens through which to ex-amine DCTT and requires hard but important work to appropriately balance competing interests within technology architecture For example there is value in technology providing users the option to collect their location history and share it with public health professionals in order to advance the public health response increase system efficiencies (eg by contributing information that can lead to better data processing) and reduce the burden on essential workers For some this might be an ex-pression of autonomy solidarity or patriotism At the same time there is value in further advancing autonomy by designing technology to allow individuals to control what data about them are collected and shared
Justifying a Middle-Ground Approach to DCTT
We ought to embrace a DCTT that has a default of interoperability and privacy protection but that does not stop there Triggering events such as entry of a positive test result or receipt of a notification that one was proximate to someone who tested positive could for example generate a push notification that users can acknowledge in order to permit transmis-sion of potentially useful location data to public health authorities This could be accompanied by an explanation of the value of the information and relevant restrictions on its use
At this point it is worth reiterating that manual contact tracingmdash
52 Digital Contact Tracing for Pandemic Response
which involves collecting information from people whorsquove tested posi-tive and their contactsmdashincludes collection of personal information and potentially embarrassing or sensitive data about the places theyrsquove been and the people theyrsquove had contact with Manual contact tracing efforts use these data to uncover ongoing transmission provide useful informa-tion tailored to the individual and enable isolation and quarantine as necessary
It stands to reason that if these forms of data can be collected by a DCTT and provided to public health authorities in a maximally secure and voluntary way (with clear rules regarding authorized uses) this may amplify public health authoritiesrsquo manual contact tracing efforts For ex-ample location data from DCTT could help jog peoplersquos memories about where theyrsquove been and fill in memory gaps This is especially relevant given the long period of infectivity of SARS-CoV-2 which begins before people are symptomatic and therefore before they are aware they are infected (Ferretti et al 2020) Location data might reveal that a COV+ person was at a restaurant at an exact time and date which could be followed up by contact tracers who could alert the public or use other measures to reach those who were also present in the restaurant at the same time In other disease contexts (see Furlanello et al 2002 Dredze et al 2013 Eckhoff and Tatem 2015 Fraccaro et al 2019) geolocation data have demonstrated some potential to support epidemiology and dis-ease surveillance with technical cautions regarding accuracy and the like (Beukenhorst et al 2017)
These benefits are currently speculative for DCTT At present pro-viding public health authorities with large amounts of data on cases and potential case contacts will be useful only if there is sufficient public health capacity to follow up on these data In addition there is a risk of low-quality data from DCTT flooding the system Investigating poten-tial case contacts identified by a DCTT may distract them from other important efforts and at some point overwhelm public health capacity altogether Whether and to what extent data from DCTT will benefit contact tracing efforts is unknown pointing again to the importance of continuously collecting high-quality evidence about DCTT
Nevertheless what would enable the most flexible and potentially robust public health response is to design DCTT so that restricted data sharing is possible From an ethics perspective the collection and use
Ethics of Designing and Using DCTT 53
of sensitive data in manual contact tracing efforts (described above) is typically seen as ethically justifiable so long as there is sufficient public health benefit and need Thus wouldnrsquot it seem appropriate from both a public health and ethics perspective to design DCTT systems to enable similar data to be shared with public health authorities when and if there is ethical justification for sharing them
Why instead do so many advocate that DCTT should be designed as a ldquominimalrdquo system when this arguably ties the hands of public health and individual users and precludes the collection of data that public health authorities (and indeed many other apps on our phones) typically collect We here consider and appraise some of the reasons that may motivate individuals and groups to argue for minimalistic positions
1 Proponents of minimal systems may believe that such systems will be most
widely adopted Some groups have maintained that only these systems will earn and maintain public trust and be widely adopted (Simpson and Conner 2020) For example the previously referenced open letter (ldquoJoint Statement on Contact Tracingrdquo 2020) asserts ldquoSome of the Bluetooth-based proposals respect the individualrsquos right to privacy whilst others would enable (via mission creep) a form of government or private sector surveillance that would catastrophically hamper trust in and acceptance of such an application by society at large It is crucial that citizens trust the applications in order to produce sufficient uptake to make a difference in tackling the crisis It is vital that in coming out of the current crisis we do not create a tool that enables large scale data collection on the population either now or at a later time Thus solutions which allow reconstructing invasive information about the population should be rejected without further discussionrdquo
Response While it is true that public trust in and acceptance of DCTT is essential for its success there is insufficient evidence that public trust would be threatened by a DCTT system that has the capacity to collect location data and enable voluntary sharing of those data with public health authorities A contrasting perspec-tive is that maintaining public trust requires maintaining public confidence that the DCTT system is providing useful information is benefiting and not harming individuals and is advancing the
54 Digital Contact Tracing for Pandemic Response
public health response (Leprince-Ringuet 2020) From this per-spective a system that is less well integrated into the broader public health response or that generates a higher rate of false positives (as some suggest decentralized approaches might (Fraser et al 2020)) may fare worse when it comes to maintaining public confidence and trust
2 Proponents may hold the view that minimal systems are harmless (or nearly
harmless) to individuals This is because individuals are anonymous none of their location data are gathered and none of their identifiable data are shared with anyone In contrast DCTT systems that collect and share identifiable data including location data may be seen as posing risks of harm to individuals
Response While minimal systems may be harmless (or nearly harm-less) from the perspective of protecting privacy they may not be harmless from the perspective of public health if they generate system inefficiencies through producing too many false positive or false negative contacts Aside from presenting a challenge for public health professionals false positives could also harm individ-uals If users receive a large volume of automated messages alert-ing them to proximity events will this cause distress Will a large volume of alerts cause users to become disengaged and stop using the DCTT or lose confidence in contact tracing more generally as a legitimate method of disease control Admittedly these are just potential harms and risks it is unknown the degree to which they will materialize The point is that privacy-related harms are not the only relevant harms to individuals that we should consider when assessing DCTT
We acknowledge the risk under a middle-ground DCTT of data being used in ethically unjustifiable and harmful ways For exam-ple it would be against the principles and recommendations artic-ulated in this report for data to be sold or monetized by technology companies or others for corporate gain and this misuse of data would be more intrusive if the data were potentially identifiable What makes it ethically justifiable to take this risk is the compen-sating benefit of allowing the most flexible and robust public health
Ethics of Designing and Using DCTT 55
response during the pandemic but this alone is not sufficient The risk of inappropriate uses must be reduced by ensuring stringent requirements for data security and access as well as clear legal protections and recourse for any violations (as discussed further below)
3 Proponents may believe that DCTT systems should not collect location data
as this would be too intrusive and of insufficient value Some proponents of PPPT systems maintain that recording proximity events is sufficient and data relating to usersrsquo movement and location should not be collected (Ingram 2020) The thought may be all we need to know is whether two individuals came into close enough contact for viral transmission to have occurred we donrsquot need to know where or when this contact occurred and there is no need to collect and store usersrsquo location data
Response This conclusion might be too hasty As discussed above there is potential (though unproven) benefit to providing public health authorities with location data Location data could help jog peoplersquos memories about where theyrsquove been provide more context for understanding the nature of ldquoproximity eventsrdquo captured by the DCTT and allow public health authorities to quickly define a category of individuals who may be at risk Collecting location data from cases is what public health authorities do on a regular basis following best practices for manual contact tracing
In addition many peoplersquos location data are currently gathered by apps on their phones and used for various purposes such as to provide more accurate navigation to offer entertainment or to improve services Many are willing to accept these capabilities because they provide some value in return Why not allow DCTT to also collect these data so that the data are available for users to share with public health officials who can then do their work more effectively and refine their understanding of how the disease transmits If many are willing to have these data used to find a bet-ter route home why not let individuals share these data to support the effort to save lives
56 Digital Contact Tracing for Pandemic Response
4 Proponents may hold the view that minimal systems pose little or no threat
to individual autonomy whereas systems that collect identifiable data and
integrate public health do pose a threat to individual autonomy For exam-ple they may worry that use of DCTT could be mandated and not a voluntary choice and in this circumstance mandatory use of minimal DCTT would be less intrusive risky and privacy violating Another worry might be that itrsquos theoretically possible that DCTT could share individualsrsquo data with public health authorities without usersrsquo full understanding if the technology does not even gather identifiable data then itrsquos not possible for these data to be shared without the individualrsquos consent
Response We discuss the importance of appropriately designed disclosures and consent below as well as the high bar that would need to be met to ethically justify mandatory use At this time mandated use of DCTT by states or institutions is not justifiable given uncertainty about potential harms and benefits Users should have a meaningful opportunity to review and understand infor-mation about the specific technology and its uses and to consent Assuming that individuals are not required to use DCTT and that they provide consent to using it designing DCTT to make data collection and sharing possible is the design choice that maximizes individual autonomy because it provides individuals with options they may value
Individuals may wish to share their data with public health au-thorities for both self-interested and altruistic reasons For exam-ple someone who has tested positive for SARS-CoV-2 and enters this test result into an app may wish to be connected to public health authorities in order to be provided with needed information resources and support She may wish for public health authorities to be provided with her phone number in case they need to reach her to provide additional information Further someone who has been alerted by an app that he had a ldquoproximity eventrdquo with a person who has tested positive for SARS-CoV-2 may wish he had location data to share with public health authorities in order to help ascertain whether this event is a cause for concern or whether it is likely a false positive (eg he and the COV+ person were sepa-
Ethics of Designing and Using DCTT 57
rated by a wall) Someone who tests positive for the virus may also wish to share their location history with public health authorities in order to be as helpful as possible to the overall public health re-sponse by facilitating de-identified aggregate analyses that identify locations of higher transmission or contribute to refining overall understanding of the disease and pandemic
5 Concerns about ldquosurveillance creeprdquo and the long-term downstream effects of
digital contact tracing system may also motivate embrace of minimal DCTT Digital contact tracing technology that collects identifiers and loca-tion data and has the capacity to share them with public health au-thorities may represent a massive and concerning increase in govern-ment surveillance of the public It might be feared that the use of this surveillance capacity in the COVID-19 response sets an unwelcome precedent for future use in other contexts Designing DCTT as min-imal systems may be a way to minimize the risk of surveillance creep and to minimize the harms associated with potential future uses of the technology
Response Surveillance creep is a serious concern To guard against surveillance creep protections should be put in place to ensure that only those data that are necessary and relevant for the public health purposes at hand are collected and used and data should be kept only for the period of time needed for those public health purposes In the face of these concerns it is important to emphasize that widespread use of DCTT in the COVID-19 response is justi-fied by the exceptional circumstances of the current pandemic and their use in this context does not imply that future public health use is ethically appropriate without significant public debate (eg use in seasonal flu surveillance efforts) Future use will require in-dependent justification Use of DCTT in other contexts (eg law enforcement or immigration enforcement) is also presumptively unethical
All in all the arguments that DCTT should be designed as a minimal system are not convincing Rather DCTT should be developed through a ldquovalues in designrdquo approach with a core set of features that protect pri-
58 Digital Contact Tracing for Pandemic Response
vacy with enough flexibility to be used differently depending upon local conditions evolving evidence and individual preferences What kind of digital contact tracing system will strike the right balance between public health goals and other considerations will depend upon circumstances For example whether it is even beneficial to provide public health au-thorities with volumes of data about potential contacts of COV+ people will depend in part upon whether they have the capacity to make good use of those data This will vary from location to location and will change over time
Recommendations
bull Technology companies should not alone control the terms condi-tions or capabilities of DCTT nor should they presume to know what may be acceptable to members of the public
bull A ldquovalues in designrdquo approach to development of DCTT should be adopted (Flanagan Howe and Nissenbaum 2008 Knobel and Bowker 2011) Robust public- and user-engagement activities should be pursued to identify and incorporate to the extent pos-sible a range of values into the design of the technology These values may include privacy but also autonomy efficiency equity or others Technology design should reflect an appropriate balance and prioritization of identified values
bull Technology design should not be static but rather it should be ca-pable of evolving depending upon local conditions new evidence and changing preferences and priorities
bull DCTT should be designed to have a base set of features that protect privacy with layers of additional capabilities that users may choose to activate An initial default should be that user location data are not shared but users should be provided with easy mechanisms and prompts to allow for opting-in to this capability with encourage-ment to the public if and as it is shown to be critical to achieving public health goals
Ethics of Designing and Using DCTT 59
Policy Positions to Advance Widespread Use of Digital Contact Tracing Technologies
The public health value of a DCTT depends in part on the number of people who use it This section concerns broad public policy positions that relate to the widespread adoption of DCTT What are ethical means of encouraging or securing widespread adoption of DCTT systems Un-der what circumstances would it be ethical to mandate their use or incen-tivize their use What enforcement challenges exist
Mandating Use
Digital contact tracing has occurred without the publicrsquos explicit volun-tary agreement in some countries such as China and Israel In others use has been voluntary (Valentino-DeVries Singer and Krolik 2020) For example Singapore adopted an app that the public could use on a vol-untary basis and approximately 20 of the population has downloaded and used it Norway has recently launched a contact tracing app that was downloaded by roughly 30 of the population in the first week that it was made available In the United States many advocates and researchers have argued that use of digital contact tracing tools must be fully volun-tary this is the dominant perspective
There are numerous ways that DCTT could be put into use without user choice For example as has been done in Israel location data from mobile phones could be collected and used by the government without usersrsquo consent Use of an app could be formally mandated as a precon-dition for returning to work or school or even further to control entry into a facility or onto transportation such as airplanes through scanning of a QR code to demonstrate personal exposure levels (Gan and Culver 2020)
While these approaches are hard to imagine in the United States some contend that mandatory use of digital contact tracing tools could be ethical and may even be ethically required Mandating use of digital contact tracing tools could in theory vastly increase the effectiveness of digital contact tracing systems and thus may save more lives and allow states to lift lockdowns sooner or avoid reimposing lockdowns in the future Canca (2020) argues that use of privacy-by-design digital contact
60 Digital Contact Tracing for Pandemic Response
tracing tools should be mandatory because the use of these tools will be nearly harmless if there are sufficient privacy protections In addition mandatory use of DCTT that embraces these principles is significantly less intrusive at the individual level than manual contact tracing which involves the collection of personally identifying and potentially sensitive data In this light it could be argued that such mandates are actually pref-erable from the perspective of both public health and individual liberty insofar as they reduce the likelihood of ldquostay at homerdquo orders which are a severe limitation of individual liberty
Nevertheless mandated use of DCTT systems faces considerable ob-stacles For example people may not adhere to the mandate by simply leaving their phone at home thus preventing their activities from being tracked Even more harmful would be if people react to a mandate and a perceived violation of liberty and privacy by employing location and Bluetooth spoofing software to shield their real contacts behind a screen of misinformation The introduction of this misinformation into a contact tracing effort might severely undermine its effectiveness The possibility of nonadherence also raises the issue of enforcement would high rates of nonadherence be permitted or would enforcement be attempted (if even possible) Perhaps more important should the technology not deliver the hoped-for benefits having mandated the use of an unproven technology could result in a loss of public trust in the technology the entity instituting the mandate and potentially the larger public health response (Bernstein et al 2019)
Mandatory DCTT could also be used to enforce quarantine restric-tions and stay-at-home orders for those who are COV+ or are determined to be at heightened risk The use of DCTT in enforcement activities raises a number of ethical (and legal) issues that are beyond the scope of the present analysis In particular individuals have a heightened interest in personal privacy if their data can be used to restrict their freedom of movement and other civil liberties At a minimum stringent procedural protections would be required to ensure that the data collection is fair and unbiased and that DCTT users are provided with adequate informa-tion in advance about how their data may be used
Mandatory use policies for DCTT must therefore convincingly ad-dress a number of questions including
Ethics of Designing and Using DCTT 61
bull Is the technology designed to meet an important and unmet public health need
bull Is there sufficient evidence to suggest that the technology will be effective at serving its purpose
bull Is the outbreak characterized by sufficiently severe morbidity and mortality and a high rate of disease transmission
bull Are there other less autonomy-restricting or less risky alternatives to widespread mandatory use of DCTT
bull Is it possible and likely that a sufficient number of individuals will comply with a mandate
bull Can inequities in the burdens and benefits of the mandate be suffi-ciently addressed through social protections and countermeasures
bull Can enforcement and enforcement discretion be implemented in a manner that is consonant with fundamental rights
bull Will those subject to the mandate interact closely with a population that is at high risk of morbidity or mortality if they contract the virus
bull Is it possible to mandate use and remain consistent with important ethical and legal principles
These questions would need to be satisfactorily addressed and explicitly documented by any decision maker considering mandatory use includ-ing government officials institutional leaders and employers Particularly important is the need to identify reliable evidence that the DCTT would be effective and to ensure that the burdens and benefits of use are equita-ble and justifiable If use of a DCTT is a condition for returning to work or school in person those who refuse or are unable to use DCTT should not lose their jobs or positions as a result and adequate support should be in place for people who are asked to self-quarantine
Finally it is important to distinguish a mandate from a ldquopushedrdquo program installation or a default setting in an application which can be modified by users A mandate relates to a policy of required use whereas the pushed programs or default settings relate to the chosen architecture for download and operation of the application
62 Digital Contact Tracing for Pandemic Response
Incentivizing Use
Perhaps the most effective way to generate widespread adoption of DCTT in the United States is to offer incentives to individuals who choose to adopt and who properly utilize the preferred DCTT approach in a voluntary system External incentives may help ldquonudgerdquo populations toward desired adoption targets Given the importance of widespread use of DCTT modest incentives ought to be considered for DCTT in the US if and when there is sufficient evidence of the technologyrsquos utility Note that in other contexts studies have shown that the provision of some incentive leads to an increase in adoption or utilization of public health programs (Singer and Ye 2013 Lee et al 2014) Moreover even a relatively small incentive can achieve much greater rates of adoption with some studies demonstrating that the incremental adoption gain de-creases as the incentive gets larger (Thornton 2008 Gibson et al 2019) In the context of COVID-19 incentives that might be both effective and ethically acceptable could include a relatively small monetary token free or discounted mobile phone service for a period of time or credit to be used by means of a mobile phone
Not all incentives are ethically appropriate For example making access to lifesaving health care contingent on using a DCTT or making valuable disease information available only to DCTT users would not be ethically appropriate In addition incentives cannot be used to over-come otherwise ethically unjustifiable technology design for example they should not be used as an offset for providing personally identifiable health information to other users
Importantly incentivization schemes must be kept distinct from man-dates as the latter require greater ethical justification To offer an incen-tive is to offer something of actual value to individual participants over and above what they are reasonably entitled to at baseline For example making a return to work contingent on using DCTT is not offering an incentive but instead imposing a mandate and it would have to be justi-fied as a mandate
In the context of COVID-19 it is also necessary to recognize that there is an inherent ldquoincentiverdquo behind the technologymdashthat is the prom-ise of more lives saved faster pandemic recovery and the reduction or elimination of blanket physical distancing Effective public communica-
Ethics of Designing and Using DCTT 63
tion of these goals if and when there is sufficient confidence in the tech-nology is important
Encouraging Use
Another important approach to increasing use of DCTT in the United States is for trusted leaders to encourage their use Community leaders public figures health care professionals and other respected individuals who have the publicrsquos trust and goodwill could be enlisted to commu-nicate with the public about DCTT and encourage its use drawing on notions such as communal responsibility solidarity and so on These en-couragements could be combined with other approaches (eg small in-centives) to optimize reach while continuing to respect individual choice
Recommendations
bull DCTT use should not be mandated at this time given uncertainty about potential harms and benefits Additional technology user and real-world testing is needed
bull Incentives can be a useful complement to encouragements how-ever any incentives for users to install and use DCTT must be equitable should not be coercive and should align with effective use of the technology (ie they should not incentivize downloading an app but then leaving onersquos phone at home)
bull Trusted leaders should be enlisted to communicate effectively with the public about DCTT and encourage its use should the technol-ogy demonstrate some potential The limits of knowledge regard-ing effectiveness should also be explained along with what will be done to improve technological capabilities as understanding evolves
Disclosure and AuthorizationConsent
In deciding whether to use DCTT voluntarily individuals must be suf-ficiently informed both through broad coordinated public engagement campaigns and individual-level disclosures and there must be a meaning-ful mechanism for users to consent It is important to recognize that while
64 Digital Contact Tracing for Pandemic Response
informed consentmdashwhich is characterized by detailed consent forms and requires a witnessed signaturemdashis the standard for most research and clinical care encounters (Faden and Beauchamp 1986) it is not typically the standard for public health disease surveillance In the public health context other relevant protections (such ethics training for public health professionals and strict data handling and confidentiality requirements) are in place and there is a strong public health interest in collecting the relevant data A more limited role for consent has been recommended for public health surveillance based on a reciprocal obligation of members of society to contribute to a ldquocommon goodrdquo and particularly in the con-text of a pandemic practical considerations such as time constraints and exigencies such as increasing morbidity and mortality (WHO 2017)
Under current circumstances given that (1) many individuals have time and capacity to consent (2) DCTT is being considered as part of plans for longer-term restabilization (3) DCTT is not a familiar part of our public lexicon (4) remote consent disclosure and authorization can be easily embedded in DCTT systems (Moore et al 2017) and (5) there are justifiable public deficits in trust with respect to various government and corporate actors handling potentially personal digital information a strong ethical case can be made for requiring a carefully crafted version of what is sometimes referred to as simple consent Simple consent consists of basic disclosure and voluntary agreement or authorization (Ali et al 2017) Three questions then arise
1 What information should be disclosed to potential users of DCTT
bull Information disclosed might include
deg Entity responsible for the technology
deg Its purpose
deg How it works (in lay terms)
Some participatory disease surveillance systems (eg Flu Near You) have received for-
mal ldquowaiversrdquo of consent requirements from institutional review boards (IRBs) in the US
As they undergo development these digital surveillance systems often straddle a line be-
tween public health surveillance and research hence the frequent need or desire to obtain
ethical review by an IRB (Ali et al 2019)
Ethics of Designing and Using DCTT 65
deg What users need to do
deg Any user options eg
Sharing geolocation data with public health authorities when that would facilitate a defined public health goal
Sharing de-identified metadata with technology develop-ers (for system enhancement)
deg User rights
deg How data will be handled
What data are collected
What data are shared (and how and with whom)
Purposes for which data can be used and not used
How data are secured and protected
Whether and what data will be retained (or will be deletable)
deg Potential benefits and any known risks
deg How to obtain answers to questions about the technology and public health response
2 How should this information be presented
Information should be presented leveraging eConsent models that are more accessible than long ldquoclickwraprdquo disclosures typical of mobile apps (Iwaya et al 2019) For example a simple open-source smartphone con-sent module that has been developed by Sage Bionetworks for research uses could be adapted to the public health surveillance context and to DCTT (Doerr Suver and Wilbanks 2016)
bull Formatting recommendations include (cf Doerr et al 2016)
deg simple and straightforward information
deg deliberately organized content
deg multimodal learning (eg visual audio written)
deg accessibility for disabled users
66 Digital Contact Tracing for Pandemic Response
deg multilingual text
deg engagement through interaction (eg swiping to navigate forward and backward)
bull The same simple information should be made publicly available via multiple other platforms (eg on websites in newspapers over social media)
bull More detailed disclosures should be made readily accessible to those who wish to learn more with no hidden surprises
3 How should users signal that they agree to the details specified in disclosures
Opt-in Models
Opt-in models are those that through an affirmative act such as clicking a button users would indicate their intention to use a DCTT This ap-proach is consistent with other app downloads where app details and privacy policies are made available through a download page and users are required to affirmatively click a button to install an app Once in-stalled some apps further alert users to particular ways in which phone capabilities or data will be used with some permitting selective toggling (opting-in or opting-out) of certain features With DCTT apps in addi-tion to disclosures provided on a download page the user could be guided through a simple interactive module embedded in the app (such as is described above) in order to increase the chance of meaningful exposure to important information about the technology and how data will be handled At that point any user options such as those itemized above could be described and choices made
Opt-out Models
There are at least two different ways in which the term ldquoopt-outrdquo has been used in this context The conventional use of the term ldquoopt-outrdquo is characterized by an act which signals an individualrsquos intention to decline something that would have otherwise occurred without intervention A few others have used the term to refer to ldquorevocation of consentrdquo for example the United States COVID-19 Consumer Data Protection Act of 2020 Senate bill (S3663) would establish a default opt-in positionmdash
Ethics of Designing and Using DCTT 67
requiring ldquoaffirmative express consentrdquo for collection and use of prox-imity and other related datamdashand refers to individuals having a right to later revoke their consent through an ldquoopt-outrdquo The latter use of the term is not our focus here
Given this a DCTT app that is voluntarily downloaded through an affirmative act would be difficult to characterize as an opt-out approach This leaves more passive surveillance systems that rely on automatic in-stallation of self-activating technology onto users phones There are a range of views among the authors of this report about the value of an opt-out approach for DCTT with some arguing for an opt-out approach on grounds that it might increase coverage and would be ethically acceptable if accompanied by similar disclosures as above to ensure users are aware of the technology and data uses (Mello and Wang 2020) This approach would present users with a mechanism to opt-out if they wish which should be reasonably easy to effectuate Under these circumstances as noted above an ldquoopt-outrdquo would not be synonymous with mandating use of the technology
Others among the authors argue that there is reason to believe that opt-in approaches may be able to sufficiently achieve desirable levels of utilization relative to opt-out approaches Unfortunately data related to opt-in versus opt-out models of DCTT are very limited One recent sur-vey (Altmann et al 2020) found that across five countries (UK Germany France Italy US) slightly more people reportedly would download an app under an opt-in system (748) than would keep an app on their phone under an opt-out system (677) Moreover when US respondents were directly asked which approach they would prefer 60 indicated a preference for opt-in This remained true across various demographic variablesmdashgender region political affiliation lockdown status and other characteristics Whether actual behaviors would align with anticipated behaviors in the context of DCTT remains an unanswered question that should be carefully studied under real-world conditions There are a range of important empirical questions regarding how much and what kind of impact (positive or negative) various types of defaults might generate for public health and for different mobile phone user groups including vul-nerable and marginalized users
Opt-out models for app authorization may encounter greater legal
68 Digital Contact Tracing for Pandemic Response
and political challenges especially if the COVID-19 Consumer Data Pro-tection Act of 2020 (S3663) the competing Public Health Emergency Pri-vacy Act (S3749) or another similar bill is enacted in the United States Both of these standing bills require affirmative opt-in consent Opt-out approaches also risk negative reactions from some mobile phone users a small number of whom may go so far as to intentionally interfere with data because of the perceived intrusiveness of an automatically installed tracking platform (Dixit 2020)
Given these considerations and the apparent willingness of a large portion of the population to opt-in to use DCTT an opt-in approach to authorization should be instituted to accompany initial DCTT rollout The feasibility and value of opt-out approaches should continue to be evaluated informed by what is technologically possible what local as-sessments of benefits and harms of the technology reveal over time and our evolving understanding of the degree to which an opt-out approach is likely to increase or decrease utilization Opt-out approaches should not be precluded
Recommendations
bull A clear and concise module consisting of basic disclosure and vol-untary authorization should be developed to accompany DCTT This module should not take the form of ldquoclickwraprdquo terms of service or end-user agreements but rather provide only essential information necessary for an individual to make a decision More detailed disclosures (such as FAQs in plain language) should be made easily accessible to those who wish to learn more with no hidden surprises
bull An opt-in approach to authorization should be instituted to ac-company initial DCTT rollout The feasibility and value of opt-out approaches should continue to be evaluated informed by what is technologically possible what local assessments of benefits and harms of the technology reveal over time and our evolving under-standing of the degree to which an opt-out approach is likely to increase or decrease utilization among different populations Opt-out approaches should not be precluded
Ethics of Designing and Using DCTT 69
Promoting Equity and Fairness
Digital contact tracing technologies should be designed and used in ways that as far as possible promote an equitable distribution of benefits and burdens DCTT should be deployed in a manner that does not propa-gate preexisting patterns of unfair disadvantage or distribute harms and risks unfairly throughout the population For example communities with lower rates of technology and data access may benefit less from DCTT Special attention must be paid to communities that experience preexisting health disparities and to those that are being hardest hit by the pandemic
Digital Disparities
In the United States February 2019 data indicate that approximately 80 of the population are smartphone users (Pew Research Center 2020) though rates of mobile phone use are significantly lower among people over age 65 (53) people with any disability (58 2016 data) (Anderson and Perrin 2017) people with less than a high school edu-cation (66) people who earn less than $30000 per year (71) and people who live in rural areas (71) As a result these populations and communities may use DCTT in lower numbers thereby lessening the effectiveness of DCTT and the likelihood of benefit for these populations from such systems Moreover it has been reported that many older and less costly smartphones (roughly estimated at 10ndash20 of smartphones in the US) lack important capabilities required for the leading AppleGoogle platform to work (Bradshaw 2020) This is of special concern because some of the above groups that are less likely to own smartphones in general are also less likely to own newer smartphones with the needed capabilities Some within the above groups (eg people who are older and people identified as Hispanic African American or American Indian) are also disproportionately experiencing morbidity and mortality from COVID-19 (CDC 2020h)
One may argue that by using DCTT human and financial resources that would otherwise be spent on manual contact tracing will be pre-served and these resources can then be redirected to better meet the needs of those who are not otherwise being effectively served by the technology because of disparities or for other reasons This argument has intuitive
70 Digital Contact Tracing for Pandemic Response
appeal and should be taken seriously however it is unsettled whether DCTT will contribute sufficient efficiencies to the overall public health response to make it possible financially and logistically for manual ser-vices to be allocated in greater proportion to those who are unable to benefit from DCTT It is entirely possible that at least in the short-term DCTT may introduce new inefficiencies due to unintended consequences or the need for public health officials to follow up many more contacts One possible mitigation to the challenge of digital disparitymdashthough it does not solve the underlying challenge of ensuring net efficiency across systemsmdashmight be to provide mobile phones or other devices and data packages to those who would otherwise be left out
Disparate Risk of Harm from Surveillance and Data Gathering
Ensuring wide digital coverage does not however resolve other equity concerns It is important to consider that some populations may experi-ence greater harm and fear of harm from having their data collected For example some groups such as African Americans Hispanic Americans Muslim Americans and undocumented immigrants have more reasonable fear of their data being handed over to law or immigration enforcement and some groups have lower levels of trust in public health due to past injustices (CSM 2017 Pew Research Center 2017 Rodrigues et al 2018 Auxier et al 2019) Any data gathered by DCTT should be used solely for public health purposes Efforts should be made to assure members of these and other communities that their data will not be misused or made available to those outside of a public health context In addition if DCTT are used in the current pandemic this should be with the understanding that future use of DCTT in other contexts (eg law enforcement or im-migration enforcement) is presumptively unethical
Some preliminary polling related specifically to DCTT emphasizes the complexity of the challenges faced and the need for deeper public engagement (Anderson and Auxier 2020) The polling results suggest that people who identify as African American or Hispanic are more likely than people who identify as White to consider government tracking of mobile phones as acceptable These findings like many others are difficult to in-terpret given background political polarization on the issue More direct engagement is required to better understand how different communities comprehend and experience DCTT and other forms of surveillance
Ethics of Designing and Using DCTT 71
Discrimination and Stigma
Stigma may result from an individual being identified as COV+ or a neighborhood or establishment becoming identified as a ldquohotspotrdquo as a result of numerous COV+ people living in that area or having visited that establishment In particular certain groups may suffer more as a result of being associated with COVID-19 such as the well-documented blame that has been directed toward Chinese people (and broadly East Asian communities) or the communities that are disproportionately likely to contract the illness (Devakumar et al 2020) When identifiable lo-cation data are made public as has been the case in South Korea per-sonal and private information were revealed Furthermore businesses in South Korea that were identified as having patrons who tested positive for COVID-19 have suffered economic losses and stigma (N Kim 2020)
To avoid the stigma and potential discrimination that can result from being identified as COV+ DCTT must never make data publicly avail-able that could be used to identify persons who have tested positive Safe-guards must be in place to ensure that any identifiable data that may be gathered for public health purposes are protected If DCTT data are used to provide heat maps to the public of locations that COV+ individuals frequently visit so as to provide representations of geographic risk or for other reasons it is essential that care be taken to avoid unfairly distrib-uting further economic burdens or other stigmatizing and discriminatory outcomes
Recommendations
bull A commitment to equity means a commitment to ensuring that the benefits and burdens of DCTT are distributed fairly Public engagement is an important tool for assessing impact and to rectify inequities
bull States localities and institutions that recommend widespread use of DCTT should provide technology (eg mobile phones Blue-tooth devices) and free data packages to those who desire but lack access to these devices
bull If there are lower rates of adoption of DCTT systems in some identifiable communities public health authorities should iden-
72 Digital Contact Tracing for Pandemic Response
tify ways to compensate For example directing more non-DCTT resources and efforts toward those communities to meet specific needs that are elsewhere being supported by technology
bull If maps are generated based on DCTT to provide the public with the locations that COV+ individuals have visited steps must be taken to minimize the stigma and potential financial losses that could result from being identified as a hotspot
Instituting Transparent Governance and Oversight
DCTT must be developed with an eye toward both present and future implications Transparent and publicly trustworthy management gover-nance and oversight of DCTT technology and data is both a near- and long-term necessity We face significant uncertainties DCTT technologies are rapidly developing Their risks capabilities effectiveness and down-stream implications are not yet well understood
Concerns about ldquoSurveillance Creeprdquo
Significant concerns have been expressed by privacy advocates (Guari-glia 2020) and in the popular press (Giglio 2020) about what is known as ldquosurveillance creeprdquo Their worry is that state and corporate actors will use new surveillance technologies capacities and permissions well beyond the purposes for which they were initially justified to the public and beyond the time when they are useful for the COVID-19 pandemic
Surveillance creep should be guarded against Only those data that are necessary and relevant for the public health purposes at hand should be collected and used and data should be kept only for the period of time needed for those public health purposes Data should be used only for public health purposes
Any use of DCTT during the current pandemic would be justified by the circumstances of this pandemic and its use in this context does not set a precedent for future public health use (eg use in seasonal flu surveillance efforts) Future use will require independent justification Use of DCTT in the future in other contexts (eg law enforcement or immi-gration enforcement) is presumptively unethical
Broadly speaking efforts should be made to generate public aware-
Ethics of Designing and Using DCTT 73
ness and consensus that use of DCTT in COVID-19 efforts does not imply that future use is justifiable However generating this public aware-ness may be particularly challenging given the complexity of the informa-tional environment where public debate ranges from legitimate concerns about surveillance creep to conspiracy theories regarding the origins of the COVID-19 pandemic (Muller 2020) This means authorities bear spe-cial obligations to be clear on how they plan to use the technologies what oversight mechanisms will be employed to address potential abuse and how they intend to publicize the conditions under which programs will be terminated making sure they are followed
Oversight and Ethical Review
We are rapidly gaining knowledge about SARS-CoV-2 and COVID-19 but we still have essential gaps in our understanding In the United States public health responses including DCTT will generally be developed and coordinated by individual states regional consortia and associations (Reston Sgueglia and Mossburg 2020) Good governance in this con-text requires transparency and the creation of oversight bodies with the appropriate expertise and representation to allow nimble and effective responses while serving as trusted representatives
To address the range of ethics-related concerns about the design and use of DCTT digital surveillance oversight committees should be estab-lished perhaps at a state level and with a platform for national coordina-tion These committees can provide ethical and regulatory review prior to and concurrent with widespread use of DCTT These committees should be composed of a diverse group of experts capable of evaluating a DCTT system locally including members of communities that experience higher rates of digital disparity
When assessing the design and use of digital contact tracing systems these committees (and the public more widely) should consider not only the risks and benefits accrued during the COVID-19 pandemic but also implications for the future What kind of precedent might use of these technologies during the current pandemic set for future use capabilities in other infectious disease outbreaks or in other social contexts (eg law enforcement) How can we navigate safe use of these technologies in a way that preserves public trust in them and enables the possibility of future beneficial use
74 Digital Contact Tracing for Pandemic Response
As a start it should be emphasized that principles offered in this and other guidance documents do not apply only during the pandemic Future efforts to advance DCTT capabilities during quieter times should make every effort to follow them
Recommendations
bull Digital surveillance oversight committees should be established expeditiously with diverse and qualified membership to provide ethical and regulatory review prior to and concurrent with wide-spread use of a DCTT system
bull Understandable and publicly accessible rules must guide the col-lection access control use storage and combination of data by government authorities public and private institutions and other parties such as public health researchers
bull Only those data that are necessary and relevant for the public health response to COVID-19 should be collected and used
bull Identifiable data should be kept only for the period of time needed for the public health response to COVID-19
bull Identifiable data collected as part of this response should not be shared with anyone other than the relevant public health authorities without additional specific informed consent of individual users
bull Before a government or institution adopts a digital contact tracing program they should state the conditions under which the digital contact tracing program will be terminated
bull Future use of DCTT to advance public health or other efforts (eg use in seasonal flu surveillance) would require independent justification DCTT designed for public health use should not be used by law or immigration enforcement
bull The principles offered in this guidance document apply both during and following the COVID-19 pandemic
75
The implementation of digital contact tracing technology (DCTT) is likely to implicate a number of US laws at both the federal and state levels This section focuses primarily on federal laws as these laws apply nationwide and generally preempt conflicting state laws A comprehen-sive assessment of the legality of any particular DCTT program would require case-specific analysis and attention to relevant state laws includ-ing any that specifically address DCTT which may soon exist in one or more states The analysis here is limited to the United States foreign and international laws will not be addressed
Many of the laws discussed in this section are privacy laws designed to protect individuals from the harms that may result from the unautho-rized or improper use of their personally identifiable information (PII) Under these laws legal concerns will generally be minimized if privacy protections are built directly into the DCTT technology (eg ldquoprivacy by designrdquo) As a general principle DCTT should be designed to collect and store only as much PII as is necessary to achieve the public health purpose Collecting only proximity data for example is likely to raise fewer legal concerns than collecting both proximity data and geolocation data Likewise creating aggregated anonymized or de-identified data will raise fewer legal concerns than using and disclosing PII
As we have argued elsewhere in this guidance document however the public health and societal crisis caused by COVID-19 may justify
Legal Considerations
FOUR
76 Digital Contact Tracing for Pandemic Response
greater encroachments on individual privacy than would otherwise be permissible Regardless of the type of data collected privacy concerns will be reduced if users are afforded the right to choose whether their PII is collected and how it is used and disclosed As such DCTT should gener-ally secure meaningful user consent before collecting PII a process which typically requires both disclosure of relevant information and agreement on the part of the user
Privacy concerns will also be reduced if the use of PII is strictly lim-ited to tracking and limiting the spread of SARS-CoV-2 The use of DCTT data for other purposesmdashsuch as commercial or law enforcement pur-posesmdashwould raise additional legal and ethical concerns In addition DCTT developers may be required to implement governance policies that ensure the secure storage of PII limit data retention periods require transparency about data sharing and maintain records of responses to data requests from government authorities
In short the legality of a DCTT program under current United States law will depend on a number of factors including what type of data is collected how the data are used and who may access them how user consent is obtained whether the entity collecting and using the data is the government or a private corporation the context in which data are collected (eg employment education or commercial) and which states have jurisdiction over the program
Privacy law in the United States unlike in other jurisdictions such as the European Union (EU) and Australia is generally sector-specific and limited in scope The result is a patchwork of protections that differ significantly depending on the entity that collects the data and the type of data collected For example under current law telecommunication carriers are governed by different privacy rules than mobile broadband providers Given the complexity of existing federal privacy law we be-lieve that it would be beneficial for the US Congress to enact new privacy legislation that is specifically tailored to the use of DCTT in response to COVID-19 Congress appears poised at least to debate such legislation a pair of bills recently introduced in the Senate and one in the House of Representatives would significantly restrict the collection of PII by digital devices for COVID-tracing purposes S3663 S3749 HR 6866 116th Cong (2020)
Legal Considerations 77
Data Privacy and Data Security Laws
Telecommunications
A DCTT provider that collects data from a userrsquos mobile phone may be subject to the privacy rules governing telecommunication carriers which are enforced by the Federal Communications Commission (FCC) The data protected under these rules are limited however to certain types of PII termed ldquocustomer proprietary network informationrdquo (CPNI) More-over the rules generally apply only to telecommunications carriers and interconnected VoIP (Voice over Internet Protocol) providers
In particular under section 222 of the Communications Act of 1934 47 USC sect 222 and the implementing regulations of the Federal Com-munications Commission (FCC) telecommunications carriers and VoIP providers must establish and maintain systems designed to ensure that they adequately protect their subscribersrsquo CPNI and they are generally restricted from using or disclosing CPNI without the customerrsquos consent (unless the use of disclosure is needed to provide the services subscribed to by the customer) If customer consent is sought to use or disclose CPNI individual notice must be provided to the customer and such notice must provide sufficient information to enable the customer to make an in-formed decision as to whether to permit the requested use or disclosure
CPNI is individually identifiable information that carriers and pro-viders have collected about their customers including phone numbers called and the frequency duration and timing of such calls Of most relevance to DCTT a recent FCC Notice of Apparent Liability asserted that user geolocation data collected by mobile phone network carriers qualify as CPNI under sect 222 and related rules 35 FCC Rcd 1785 (2) (2020) Pursuant to this notice the FCC fined T-Mobile for selling to third parties location data that were derived from the communication between the mobile phones of T-Mobilersquos customers and nearby network signal towers (The FCC also levied fines against ATampT Verizon and Sprint on the same grounds (Valentino-DeVries 2020)) While the FCC has made its position clear that geolocation data are CPNI courts have yet to weigh in on the matter
Even if geolocation data are CPNI however the FCC can enforce sect 222 of the Communications Act only against telecom carriers and VoIP
78 Digital Contact Tracing for Pandemic Response
providers not against cable broadband and mobile broadband internet providers 47 USC sect 53(44) 47 CFR sect 93 In 2018 the FCC promul-gated a regulation stating that contrary to its prior position its sect 222 authority does not extend to cable broadband and mobile broadband internet providers Restoring Internet Freedom 83 Fed Reg 7852 (Feb 2 2020) (to be codified at 47 CFR pts 1 8 and 20) This regulatory shift was subsequently upheld by the DC Circuit Mozilla Corporation v Federal Communications Commission 940 F3d 1 (2019)
In addition to sect 222 the FCC has authority to regulate ldquocommon carriersrdquomdashincluding both telecommunication carriers and broadband internet providersmdashunder sect 201(b) of the Communications Act In the past the FCC has interpreted sect 201(b) to protect against ldquounjust and unreasonablerdquo privacy and data security practices with respect to custom-ersrsquo personal information beyond CPNI In 2016 the FCC promulgated a regulation asserting its authority under this interpretation However Congress overturned this regulation pursuant to the Congressional Re-view Act in 2017 SJ Res 34 115th Cong (2017) At present the extent of the FCCrsquos authority under sect 201(b) remains unsettled (Mulligan and Linebaugh 2019)
Consumer Protection
The collection storage release and transmission of digital user data in-cluding proximity contacts is more generally governed by the Federal Trade Commission (FTC) The FTC is an independent US law enforce-ment agency tasked with protecting consumers and promoting competi-tion across broad sectors of the economy (FTC 2020) The FTCrsquos primary legal authority with respect to consumer protection comes from Section 5 of the FTC Act which prohibits ldquounfair or deceptive acts or practices in or affecting commercerdquo 15 USC sect 45(a)(1) Note that the FTC and FCC have some overlapping authority to protect consumer privacy in the context of telecommunications (FCC and FTC 2017)
The FTC has interpreted Section 5 to require companies to be trans-parent and accurate about their collection of PII from consumers A com-pany may be found to have engaged in a deceptive practice if it fails to disclose that it is collecting user data or fails to disclose that it is sharing these data with third parties and to provide a general description of these third parties The FTC has used its authority under Section 5 numerous
Legal Considerations 79
times to discipline companies that purport in published privacy policies or other notices to provide protection for the privacy andor security of personal information yet fail to do so in practice For example the FTC may find it both ldquounfairrdquo and ldquodeceptiverdquo for a mobile app privacy policy to state that the app never discloses location information to third parties when in fact the app shares that information with the app developerrsquos service provider which in turn uses it to provide analytical data to the app developer that are used to create targeted advertising
The FTC does not use its Section 5 authority other than to protect consumers and generally does not consider ldquode-identifiedrdquo user data which are data that are not ldquoreasonably linkablerdquo to a consumer to be a subject for consumer protection In general data collected are not ldquorea-sonably linkablerdquo so long as the company collecting it ldquo(1) takes rea-sonable measures to ensure that the data are de-identified (2) publicly commits not to try to reidentify the data and (3) contractually prohibits downstream recipients from trying to reidentify the datardquo (FTC 2012)
Many states have laws that are similar to Section 5 prohibiting un-fair and deceptive acts and practices Both Section 5 and these similar state laws can be violated not only by misrepresentations (affirmative deception) but also by material omissions Thus a failure to inform an app user of the apprsquos collection of tracking data and the planned use and disclosure of those data could constitute a violation of these laws Com-panies providing DCTT apps should make sure that all such information is disclosed in the appsrsquo terms of use to which users must affirmatively agree
Childrenrsquos Online Privacy
Children who use DCTT may be protected by additional privacy protec-tions In particular collection of digital PII from children under the age of 13 is strictly regulated under the Childrenrsquos Online Privacy Protection Act (COPPA) (15 USC sectsect 6501ndash6505) Under COPPA PII includes ldquofirst and last name[] a persistent identifier that can be used to recognize a user over time and across different online services[] and geolocation infor-mation sufficient to identify street name and name of a city or town[]rdquo COPPA prohibits a website or online service from collecting personal information (including location information) from children under age 13 without obtaining verifiable parental consent Note that there may be an
80 Digital Contact Tracing for Pandemic Response
exception to this requirement for an ldquoinvestigation on a matter related to public safetyrdquo 16 CFR sect 3125(c)(6)(iv)
Electronic Surveillance
In addition to misuse of user data by DCTT providers another privacy concern is that a third party may be able to access sensitive PII that is collected and stored by a DCTT system without the userrsquos knowledge and consent There are a number of federal criminal laws however that would likely prohibit such unauthorized access to PII
In particular the Electronic Communications Privacy Act of 1986 (ECPA)mdashwhich includes the Wiretap Act (18 USC sectsect 2510ndash2522) the Stored Communications Act (18 USC sectsect 2701ndash2711) and the Pen Register Act (18 USC sectsect 3121ndash3127)mdashmakes it a crime to access elec-tronic communications without authorization Individuals who violate the ECPA face up to five years in prison and fines up to $250000 Victims are also entitled to bring civil suits and recover actual damages in addi-tion to punitive damages and attorneyrsquos fees for violations
Generally the access restrictions in the ECPA apply unless consent is given or if access is authorized by statute for law enforcement purposes For example an employer is generally forbidden from accessing an em-ployeersquos private emails However if consent is given in the form of an employment contract that explicitly authorizes the employer to access emails it may be lawful under the ECPA for the employer to access such information Along the same lines the ECPA would likely prohibit an employer from accessing contact tracing data on an employeersquos phone without the employeersquos consent However the ECPA would likely not prohibit duly authorized government public health officials from access-ing contact tracing data without consent
As its name suggests the Stored Communications Act (SCA) regu-lates access to communications at rest that is not in transit The SCA makes it unlawful to intentionally access a facility in which electronic communication services are provided and to obtain alter or prevent au-thorized access to a wire or electronic communication while it is in elec-tronic storage in such a system As such the SCA would likely apply only to centralized collection of contact tracing data
The Pen Register Act covers any ldquosignaling informationrdquo exchanged in a communication such as phone numbers The statute does not reach
Legal Considerations 81
the content of such communications however An expansive interpreta-tion of the Pen Register Act would cover Bluetooth ldquohandshakesrdquo as they are merely signaling information between devices which do not carry content See United States v Forrester 512 F3d 500 (9th Cir 2007) (find-ing that email headers and IP addresses are akin to pen registers and have no Fourth Amendment protection) Unlike the SCA there is no statutory exclusionary rule that applies when the government illegally uses a pen register trap and trace device Additionally there is no private cause of action against the government for violations of the Pen Register Act
State Data Privacy Laws
States have a variety of privacy laws and are increasingly seeking to reg-ulate the online collection of personal information and the use and dis-closure of such information To date most of these laws focus more on transparency and protection from unauthorized access than on restricted collection and use (except with respect to biometric information) seek-ing to ensure that individuals who use websites or online services such as mobile applications do so on an informed basis with respect to the privacy provided by those sites and services Two examples of such state laws are the California Online Privacy Protection Act (CalOPPA) and the California Consumer Privacy Act (CCPA) Both laws require notice to in-dividuals who use websites or online services such as mobile applications in order to ensure that users are informed about the privacy of personal information collected by those sites and services (The CCPA also applies to data collection off-line) Both laws treat IP addresses and location data as types of potentially identifiable personal data and so would very likely apply to DCTT apps used by California residents
CalOPPA requires that the operator of any website mobile appli-cation or other online service (ldquoSiterdquo) post a privacy policy on the Site disclosing certain information regarding the Sitersquos collection use and dis-closure of PII CalOPPA applies to any Site that is accessible to California residents The required disclosures are not onerous and would apply only to collection of data that are identifiable to an individual person (but depending on who collects the data location data together with a device identifier are identifiable to the user)
The CCPA requires that any entity qualifying as a ldquobusinessrdquo provide its ldquoconsumersrdquomdashdefined as lawful residents of Californiamdashwith specific
82 Digital Contact Tracing for Pandemic Response
disclosures about the businessrsquos collection use and disclosure of personal information Importantly the CCPA applies only to for-profit businesses that meet certain thresholds of revenue or access to consumer informa-tion A public health agency or a nonprofit organization would not be subject to the CCPA Cal Civ Code sect 1798140(c)
The CCPA defines ldquopersonal informationrdquo as ldquoinformation that iden-tifies relates to describes is reasonably capable of being associated with or could reasonably be linked directly or indirectly with a particular consumer or householdrdquo The statute provides a nonexclusive list of po-tential identifiable personal information including ldquogeolocation datardquo In accordance with the CCPA businesses must provide consumers with a notice ldquoat or before the point of collectionrdquo of personal information which must describe the personal information to be collected and the pur-poses for collecting that information Businesses must additionally allow consumers to request access to and request deletion of personal informa-tion Businesses must allow for consumers to opt-out of the sale of any personal information Developers of COVID-tracing apps would want to build in compliance with these requirements In addition California Civil Code sect 1798815(a)(1) requires companies to ldquomaintain reason-able security procedures and practices appropriate to the nature of the information it processesrdquo
Like privacy laws generally the CCPA does not grant consumers rights regarding the use of de-identified information However the CCPA does require businesses to implement processes that prohibit re-identification of de-identified information as well as technical safeguards to prevent inadvertent release of that information Cal Civ Code sect 1798140(h)
Health Information Privacy
Many DCTT systems will be designed to collect health-related data of users such as symptom tracking SARS-CoV-2 test results and prior ex-posure to a person who is COV+ Individuals may have additional privacy protections with respect to the use and disclosure of this health-related information
The use and disclosure of individually identifiable health information is strictly regulated under the privacy and security rules implementing the
Legal Considerations 83
Health Insurance Portability and Accountability Act (HIPAA) HIPAA is limited in application however to health care providers and health insur-ance plans (ldquocovered entitiesrdquo) and ldquobusiness associatesrdquo of such entities ldquoBusiness associatesrdquo under HIPAA are persons who perform services for covered entities and need access to personal health information to do so
HIPAA-covered entities must have written authorization to use or disclose identifiable health information (ldquoprotected health informationrdquo or PHI) from the individual to whom such information pertains unless the HIPAA regulations promulgated by the US Department of Health and Human Services (HHS) provide an exception to the requirement for such individual authorization
Among the exceptions to the individual authorization requirement is an exception for certain uses and disclosures of PHI for public health purposes 45 CFR sect 164512(b) This exception would permit for exam-ple a HIPAA-covered entity to disclose the PHI of an individual who tests positive for SARS-CoV-2 to a public health authority A ldquopublic health authorityrdquo is an agency or authority of the US government a state ter-ritory a political subdivision of a state or territory or Indian tribe that is responsible for public health matters as part of its official mandate as well as a person or entity acting under a grant of authority from or under a contract with a public health agency such as a contact tracer Id sect 164501
Many DCTT developers are HIPAA business associates and any use and disclosure of PHI collected through DCTT used on behalf of HI-PAA-covered entities is restricted under the HIPAA privacy rules Nota-bly in response to COVID-19 HHS announced that its Office for Civil Rights would exercise its enforcement discretion and would not impose penalties for violations of certain provisions of the HIPAA Privacy Rule against health care providers or their business associates for the good faith uses and disclosures of protected health information for public health and health oversight activities during the nationwide public health emergency 85 FR 19392 (2020)
Many states also have health information privacy laws The HIPAA privacy rule sets a ldquofloorrdquo of privacy protections allowing the states to be more protective of privacy More specifically HIPAA preempts a state law if (but only if) the state law is ldquocontraryrdquo and less protective of privacy than the HIPAA privacy rule However if a state law is determined by the
84 Digital Contact Tracing for Pandemic Response
Secretary of HSS to be necessary to serve a ldquocompelling need related to public health safety or welfarerdquo it may survive preemption even if it is less privacy-protective than HIPAA 45 CFR sect 160203 (a)(1)(iv)
The Public Health Service Act also restricts the use of certain per-sonally identifiable information collected by entities involved with public health activities without the individualrsquos consent 42 USC 242m(d)
Labor and Employment Privacy Rights
Labor and employment lawsmdashthat is laws that govern the relationships between employers and employeesmdashmay prove relevant to DCTT espe-cially if employers mandate the use of DCTT or seek to collect health information regarding their employees using DCTT Depending on the built-in privacy protections of the DCTT system an employer may be able to access important health information from an employeersquos phone As noted above the ECPA would generally prohibit an employer from ac-cessing this information without the employeersquos consent Even with con-sent however there are limits on the collection and use of an employeersquos health information
In particular the use of DCTT may raise special concerns about em-ployment discrimination for example if an employer were to fire an em-ployee who tests positive for SARS-CoV-2 (COV+) or who has a known SARS-CoV-2 exposure The Americans with Disabilities Act (ADA) pro-tects disabled employees from discrimination and restricts the collection of personal health information by employers The Equal Employment Opportunity Commission (EEOC) which is the federal agency tasked with enforcing the ADA in the employment context would likely con-sider COV+ to be a ldquodisabilityrdquo under the ADA and analogous state laws prohibiting discrimination against disabled people COV+ is likely to be a ldquodisabilityrdquo especially where the individual is symptomatic andor experi-ences related health issues or if it is later determined that testing positive for SARS-CoV-2 leads to long-term or chronic health effects ldquoExposure to a COV+ personrdquo could also be covered by those laws because a person exposed to a COV+ individual could well be perceived as being disabled by being considered likely to be infected
Legal Considerations 85
The ADA generally requires that businesses make ldquoreasonable accom-modationsrdquo for persons who are disabled which may include individuals who are COV+ or who have a preexisting disability that places them at higher risk from or may be exacerbated by COVID-19 The EEOC has published guidance on reasonable accommodations under the ADA and related laws in the context of COVID-19 (EEOC 2020) Among other things this guidance clarifies that consistent with the ADA employers may take temperatures or otherwise collect health information about employees during the pandemic crisis so long as they keep that infor-mation confidential As of May 18 2020 the EEOC has not provided guidance that specifically addresses the applicability of the ADA to the use of DCTT by employers
In addition employment laws such as the ADA and the Family and Medical Leave Act (FMLA) and state law equivalents generally limit disclosure of information and require employers to keep confidential any employee personal health information related to a disability or request for medical leave Under the ADA any information regarding the medical condition or history of an employee that an employer obtains as part of an examination or inquiry into a disability could constitute a confidential medical record that can be disclosed only to certain individuals in lim-ited circumstances 42 USC sectsect 12112(d)(3)(B) and 12112(d)(4) The FMLA also prevents the disclosure of records related to medical histories in connection with an employeersquos leave request or eligibility 29 CFR sect 825500(g) The EEOC and some courts have gone further and taken the position that any information concerning an employeersquos medical con-dition is protected under the ADA or FMLA
As discussed elsewhere in this guidance document employers may have a good reason to employ DCTT in order to ensure workplace safety and limit the spread of SARS-CoV-2 in the community Employers may also face legal liability if they fail to protect employees (or customers) from potential exposure or infection In particular employers have an obligation under the Occupational Safety and Health Act to keep the workplace safe for employees In response to COVID-19 the Occupa-tional Safety and Health Administration (OSHA) has developed guid-ance on preparing the workplace (OSHA 2020) The CDC has also pre-pared guidance on healthy business operations and reducing the spread
86 Digital Contact Tracing for Pandemic Response
of SARS-CoV-2 in the workplace (CDC 2020c) Employers must strike an appropriate balance between avoiding employment discrimination and promoting workplace safety
Reflecting the need for such a balance the employee protections un-der the ADA and other employment laws are not absolute and are limited by among other things the need to protect the health and safety of other employees and the public Protection for workplace safety and health generally will justify appropriately tailored measures such as inquiries into an employeersquos personal health status or whether someone has tested positive for SARS-CoV-2 temperature checks and removal of employees from the workplace who are experiencing symptoms or have tested posi-tive and have not been cleared to return to work
Note finally that the use of DCTT by employers should be evaluated in conjunction with the hazard pay sick leave and other benefits that are available to employees Under the Families First Coronavirus Response Act employers with more than 50 employees and fewer than 500 employ-ees are required to provide two weeks of paid sick leave to an employee who stays home because of COVID-19 Pub L No 116-127 134 Stat 178 (2020) This paid leave extends to those who are themselves ill are quarantined or are awaiting a diagnosis as well as those who are caring for sick family members However reporting suggests that more than 75 of US workers will not qualify for benefits under this act (Cochrane Miller and Tankersley 2020)
Constitutional Privacy Rights
A DCTT program involving only private actors operating on the ba-sis of voluntarily provided information would not present constitutional privacy issues But any government-directed use of digital technology to support public health tracking and contact tracing involving mandatory government surveillance may potentially implicate a variety of consti-tutional protections These constitutional protections apply to actions taken by any level of government in the United States While state gov-ernments have broad policing powers in the area of public health (Jacob-son v Massachusetts 197 US 11 (1905)) and are generally allowed to enforce legislation not preempted by federal laws even emergency and
Legal Considerations 87
health-protective laws must be consistent with the US Constitution (HHS 2019 CDC 2020f)
Fourth Amendment Search and Seizure
Many people considering whether to use a DCTT app may be concerned that government enforcement agencies would obtain tracing data and use those data to conduct criminal prosecutions or immigration proceed-ings Constitutional protections notably the Fourth Amendmentrsquos limit on warrantless searches limit the governmentrsquos use of personal data in the criminal context However exceptions exist allowing law enforce-ment to access information even when such access would generally be prohibited How the government accesses personal data stemming from contact tracing needs to be scrutinized and protections will hinge on the manner of access
In general the Fourth Amendment protects ldquo[t]he right of the peo-ple to be secure in their persons houses papers and effects against un-reasonable searches and seizuresrdquo As originally interpreted the Fourth Amendment was considered tied to common-law trespass That is no lon-ger the case US Supreme Court precedent interprets the Fourth Amend-ment to protect ldquopeople not placesrdquo and extends to the protection of certain expectations of privacy such as location information as long as such expectations are reasonable Katz v United States 389 US 347 351 (1967) A warrantless government search is unconstitutional when the information sought is private and such expectation of privacy is ldquoone that society is prepared to recognize as reasonablerdquo Smith v Maryland 442 US 735 743ndash44 (1979)
The constitutionality of a search will revolve around the following analysis whether the digital program either violates an individualrsquos ldquorea-sonable expectation of privacyrdquo (likely triggered by programs collecting large amounts of location andor health data) or involves a government ldquotrespassrdquo (likely triggered by required app downloads) Katz v United States 389 US 347 (1967) United States v Jones 565 US 400 (2012)
Courts will most likely weigh the intrusiveness of the measures taken in implementing a search standard against the severity of the situation governmental and individual interests and accountability measures and safeguards built into the system
Voluntary sharing by individuals of their information with other par-
88 Digital Contact Tracing for Pandemic Response
ties including the government would mean that there was no reasonable expectation of privacy and would not raise the issues elaborated above It is worth noting that consent may not be considered voluntary if coerced or conditioned especially with regard to public employees or students of public institutions
Third-Party Doctrine
Some legal doctrines allow for the governmentrsquos acquisition of otherwise private information consistent with Fourth Amendment privacy protec-tions The third-party doctrine for example provides that individuals have no reasonable expectation of privacy in information voluntarily shared with others even if the information is revealed on the assumption that it will be used only for a limited purpose and the confidence placed in the third party will not be betrayed Smith v Maryland 442 US 735 (1979) United States v Miller 425 US 435 (1976) This applies to in-formation provided by third parties (mobile carriers internet service pro-viders medical tracking device manufacturers etc) to the government under order or request even when the third partyrsquos end-user agreements or privacy policies create an expectation of privacy
The Supreme Court has narrowed the applicability of the third-party doctrine to exclude use and disclosure of ldquohistoricalrdquo cell-site location information (CSLI) data For example in Carpenter v United States 138 S Ct 2206 (2018) the Court reasoned that the third-party doctrine does not justify use and disclosure of historical CSLI because an individual does not provide that information voluntarily Rather that information is pervasively collected by the cell phone company without any affirmative action on the part of the individual The Court did not express a view on ldquoreal-timerdquo CSLImdashlocation information that live-tracks a cell phonersquos locationmdashor on GPS data that may be stored in the phone itself
The Carpenter decision builds on a line of cases related to searches of digitally stored location data In Riley v California 134 S Ct 2473 (2014) the Court held that absent exigent circumstances law enforce-ment must obtain a warrant to search an individualrsquos phone Exigent circumstances are those that require immediate action because there is a probability that evidence may be destroyed The use of a centralized data-base for collection of digital contact tracing data would obviate deletion
Legal Considerations 89
concerns If the data are stored locally in the phone issues may arise as to whether law enforcement may suspect the data may be deleted following an arrest
Similarly in United States v Jones 132 S Ct 945 (2012) Justice So-nia Sotomayor authored a concurring opinion arguing that the use of a GPS to track a defendantrsquos whereabouts has the potential of providing the government with enough data points to create a ldquomosaicrdquo of the personrsquos life Location data obtained through centralized location contact tracing have the potential of providing information on an individualrsquos where-abouts beyond whatrsquos necessary for determining proximity to infected individuals Localized data may also raise the same issues if accessed by law enforcement
Following Carpenter several courts have addressed the constitution-ality of novel location tracking In Massachusetts for instance a federal district court concluded that police use of a ldquopole camerardquo on a utility pole to investigate the movements of an individual constituted a search under the Fourth Amendment United States v Moore-Bush 381 FSupp3d 139 (D Mass 2019) The court reasoned that even in a public space an in-dividual still retains a reasonable expectation of privacy ldquoin the whole of their physical movementsrdquo Citing Carpenter and Jones the court stated that the governmentrsquos unrestrained power to collect data that reveal pri-vate aspects of identity is susceptible to abuse and gives police access to a category of information that is ldquootherwise unknowablerdquo Long-term monitoring of a personrsquos movements consequently violates that individ-ualrsquos expectation of privacy Notably the court emphasized the capability of the camera to create a searchable digital log of the photos taken for the eight-month period during which the camera was used
State courts have also weighed in on the issue The Massachusetts Supreme Judicial Court found that police access to real-time location data pinpointing an individualrsquos movement whether from a third party or a cell-site simulator infringes upon an individualrsquos reasonable expec-tation of privacy Commonwealth v Almonor 120 NE3d 1183 1195 (Mass 2019) The Washington Supreme Court for its part held that a cell phone ping used to locate the defendantrsquos vehicle in real time is a search under the Fourth Amendment requiring a warrant absent exigent circumstances State v Muhammad 428 P3d 1177 (2018) And the Colo-
90 Digital Contact Tracing for Pandemic Response
rado Court of Appeals held that police use of a video pole camera to con-tinuously surveil a defendantrsquos fenced-in backyard constitutes a search under the Fourth Amendment People v Tafoya 2019 BL 457321 Colo Ct App 17CA1243 (2019)
Application of Carpenter by lower courts to novel location-tracking tactics is still evolving and it is as yet unclear how the narrower interpre-tation of the third-party doctrine will continue to be expanded and ap-plied particularly in cases of short-term monitoring of massive amounts of location andor health data Moreover it is unclear whether Carpenter would apply to DCTT data collected by the government itself
Special Needs Doctrine
An argument in favor of the constitutionality of government DCTT programs is that the ldquospecial needsrdquo doctrine would apply Under this doctrine a warrantless search that would otherwise violate the Fourth Amendment might be permissible based on a special need relating to pub-lic health When the search is conducted for a nontraditional law enforce-ment purpose and circumstances make securing a warrant impracticable the Supreme Court has ruled that warrantless searches may be permissi-ble The special needs doctrine however is highly controversial because it is not a consistently applied Fourth Amendment exception so it is diffi-cult to predict when courts would authorize nontraditional surveillance Some factors considered by the court are (1) the balance between the intrusiveness of the government action and the anticipated public bene-fits (2) the existence of legislative authorization (3) judicial process or the ability of the subject individual to challenge the government action (4) the scope or breadth of government action and (5) the likelihood of the collected data being used in criminal proceedings The Supreme Court did note in Chandler v Miller 520 US 305 (1997) that a ldquorisk to public safety [that] is substantial and realrdquo may justify ldquoblanket suspicionless searches calibrated to the riskrdquo citing as examples the routine searches conducted at airports and entrances to some official buildings (Searches within the context of immigration are further analyzed below)
Immigration Enforcement
Exceptions apply to the constitutional requirement that a warrant ac-company an unreasonable search or seizure in the immigration context
Legal Considerations 91
For example an exception to the general warrant requirement is the bor-der search exception which allows government officials to search and seize without a warrant persons and property at the border or at the functional equivalent of a border See United States v Montoya de Her-nandez 473 US 531 (1985) United States v Flores-Montano 541 US 149 (2004) Federal regulation authorizes immigration officials to oper-ate within 100 miles of any US external boundary (See 8 CFR sect 2871 defining ldquoreasonable distancerdquo as ldquowithin 100 air miles from any external boundary of the United Statesrdquo) A functional equivalent of a border may include any airport where international flights may be received automo-bile checkpoints servicing international traffic and vessels in territorial waters Government officials however must still have ldquoreasonable suspi-cionrdquo of an immigration violation or a crime to search or seize persons or property
In the context of digital data Customs and Border Protection (CBP) officials may conduct either manual or forensic searches of electronic devices at the border or its functional equivalent A manual search is considered a routine search and may include accessing the phone and ldquobrowsingrdquo its contents If the electronic device is password protected individuals must provide information for unlocking the device Forensic searches on the other hand are nonroutine and involve a more invasive search of the electronic devicersquos contents Federal circuit courts are split on whether a CBP agent needs ldquoreasonable suspicionrdquo before conducting a forensic search of an electronic device But Supreme Court precedent clearly states that suspicionless searches are not unconstitutional when public safety is considered Skinner v Ry Labor Execsrsquo Assrsquon 489 US 602 (1989)
A recent CBP directive provides guidance and standard operating procedures regarding forensic searches of electronic devices CBP 3340-049A Border Search of Elec Devices (DHS 2018) The directive states that CBP officers may detain electronic devices or copies of the informa-tion contained within these devices for a reasonable period time not to exceed five days This directive raises the concern that travelers may be required to turn over contact tracing data stored on their phone to CBP officers Note that the directive has been challenged in federal court and is currently awaiting appeal Alasaad v Nielsen 419 FSupp3d 142 (D Mass 2019)
92 Digital Contact Tracing for Pandemic Response
Searches in Schools
Another exception to the general warrant requirement applies to searches by non-law-enforcement government officials in public schools (ie school officials) Within this context school officials have broad powers to conduct searches as long as those searches are reasonable Searches by individuals in private schools are not governed by the Fourth Amend-ment State regulation of searches in private schools varies (See US DOE 2009)
Related Federal Privacy Statutes
Outside the Fourth Amendment context certain laws provide protections against government collection of and access to personal data The USA Freedom Act of 2015 for example bans the governmentrsquos bulk collec-tion of internet metadata and telephonic records which was previously allowed under Section 215 of the USA Patriot Act The government must now identify with specificity the identity of a person account address or personal device when requesting records The law allows for the acqui-sition of data by two degrees of separationmdashor ldquohopsrdquomdashfrom targeted individuals If a centralized system in contact tracing is used it is unclear whether the government may need to resort to this provision since it would likely have consent from individuals to collect and use the data
The Privacy Act of 1974 also regulates the collection use and disclo-sure of personal data but applies only to federal agencies (and their con-tractors) not to state or local agencies 5 USC sect 552a The Act protects against disclosure of individually identifying ldquorecord[s]rdquo that are kept within a ldquosystem of recordsrdquo The Act limits disclosure of information ldquoexcept pursuant to a written request by or with prior written consent of the individual to whom the record pertainsrdquo Certain disclosures are ex-empt from the Actrsquos applicability Pertinent disclosure exceptions are for records required to be disclosed under the Freedom of Information Act (FOIA) or disclosures ldquoto a person pursuant to a showing of compelling circumstances affecting the health or safety of an individualrdquo A disclosure under FOIA however would not include information in ldquopersonnel and medical files and similar filesrdquo when disclosure ldquowould constitute a clearly unwarranted invasion of personal privacyrdquo FOIA Guide 2004 Edition
Legal Considerations 93
Exemption 6 If non-anonymized data are turned over to the federal or state governments it is important to consider whether PII would be pro-tected from disclosure under FOIA or state freedom of information laws
Consent
User consent is a cross-cutting issue for evaluating many of the laws and regulations governing personal information privacy discussed in the prior sections In general privacy laws can be justified on the grounds that an individual should have the option to control with various types and de-grees of limitation the collection use retention andor disclosure of in-formation pertaining to that individual by others As such many privacy laws start from the premise that absent an individualrsquos consent use or disclosure of that individualrsquos PII is impermissible except for certain enu-merated purposes deemed to outweigh the individualrsquos privacy interests
Consent like agreements in general can be manifest in different ways in specific circumstances In some cases an affirmative actionmdashsuch as a signaturemdashis needed to demonstrate consent In other cases inactionmdashsuch as declining to ldquounsubscriberdquo from receiving certain unsolicited emailsmdashconstitutes consent Where a law requires a ldquowrittenrdquo signature in all but a few contexts the signature may be executed electronically In the United States that means the ldquosignaturerdquo may consist of any of the following ldquoan electronic sound symbol or processrdquo so long as it is ldquoattached to or logically associated with a contract or other record and executed or adopted by a person with the intent to sign the recordrdquo Electronic Signatures in Global and National Commerce Act (E-SIGN) (15 USC 7006)
The scope of a consent depends on what was deemed to be under-stood by the consenting party That is least clear when the consent is in-ferred from inactivity even if terms stating the consequences of inactivity have been provided The scope of consent is most clear when the terms agreed to have been presented to or provided by the consenting party in a conspicuous documented manner and a record exists of those terms Courts uphold the validity of clickthrough agreements because users are deemed to review the terms to which they respond by clicking ldquoI agreerdquo
94 Digital Contact Tracing for Pandemic Response
But where terms are ambiguous or confusing buried in other text or presented obscurely the ldquoI agreerdquo action may not mean the user actually agreed to specific terms
Terms of Use and Privacy Policies for apps are often written in com-plicated or nuanced language with key points difficult to discern More-over they are generally hard to read on a mobile device Many users of mobile phone apps agree to such terms without even attempting to read or to understand them As such it is often questionable whether an app user has knowingly agreed to all the terms of those documents Presen-tation of terms in large typeface short sentences simple language and direct disclosures makes user consent more meaningful
For contact tracing apps that collect PII andor PHI consent will overcome the restrictions of many if not most privacy laws provided the consent is freely given reflects a full understanding of the terms for use collection and disclosure of the information and is confirmed by an af-firmative act such as a click that may be executed only upon a complete reading of Terms of Use and Privacy Policies Whether consent may be deemed ldquofreely givenrdquo in certain circumstances depends on contextual understandings of party relationships including the employer-employee and government-citizen relationships
Anti-discrimination and Individual Freedom Laws
Any measure taken to protect public health and safety must comply with the Constitution and civil rights laws such as the ADA that prohibit discrimination against persons in certain protected categories such as race gender religion or disability In addition certain implementations of DCTT could be challenged under a variety of individual freedom protections
Anti-discrimination Laws
In general it would be impermissible to use DCTT in a way that either targets or excludes people on the basis of their membership in one of these protected categories
When motivated by animus against a protected class as defined by law and not narrowly tailored to advance a compelling government inter-
Legal Considerations 95
est a discriminatory regulation would be considered unconstitutional un-der the Equal Protection Clause of the Fourteenth Amendment to the US Constitution See Jew Ho v Williamson 103 F10 CCND Cal (1900) (striking down a quarantine imposed by San Francisco in response to an outbreak of bubonic plague because it was racially motivated) see also Church of Lukumi Babalu Aye v Hialeah 508 US 520 (1993) (supposed public health measure unconstitutional because it targeted the practices of one religion)
The risk of unintentional yet illegal discrimination in using DCTT is a real possibility Recent studies of infection rates among the population have revealed a larger-than-proportional infection rate among certain minority communities such as Latinx African American and American Indian communities (NYC DOH 2020) Programs that target specific ra-cial ethnic tribal or religious groups may raise constitutional and other legal concerns
Religious Freedom Laws
The use of DCTT may also raise concerns about religious freedom For example there may be religious objections to restrictions on gathering for worship carrying a mobile phone or the use of imaging technology Under current Supreme Court precedent generally applicable laws that do not discriminate against religion on their face do not violate the Free Exercise Clause of the First Amendment even if those laws have an inci-dental effect on the exercise of religion Employment Div Dept of Hu-man Resources of Oregon v Smith 494 US 872 (1990) These laws need not be justified by compelling government interest (the ldquostrict scrutinyrdquo standard of review) the government need only show that they are ratio-nally related to a legitimate interest On the other hand laws that are not neutral and not of general applicability must be justified by compelling government interest and must be narrowly tailored to advance that in-terest if it burdens religious practicesmdasha very tough hurdle to overcome Church of Lukumi Babalu Aye v Hialeah 508 US 520 (1993)
This general approach however is disrupted in some contexts by statutes adopted to provide greater protection to religious freedom The federal Religious Freedom Restoration Act (RFRA) requires strict scru-tiny for federal actions that burden religion and many states have ad-opted ldquostate RFRAsrdquo that do the same for actions by state and local
96 Digital Contact Tracing for Pandemic Response
governments The Religious Land Use and Institutionalized Persons Act which extends similar protections to persons confined to an institution such as a prison jail or mental health facility may also be relevant 42 USC sect 2000cc
Under either standard of review courts will examine whether a gov-ernment action imposes a substantial burden on religious exercise if not no religious freedom violation has occurred Such a finding is unlikely for DCTT programs absent evidence that the government is using the digital information to take action against religious persons that is not necessary to avoid the spread of a serious disease Nevertheless legal challenges on religious freedom grounds cannot be ruled out
Legislative Recommendations
bull Congress should enact new legislation specifically tailored to fa-cilitate the use of DCTT as part of the public health response to COVID-19 while also protecting user privacy and ensuring data security
bull Congress should require DCTT developers to disclose to users in clear language the nature of the information that would be col-lected how it would be collected how it would be stored and for what purposes it may be used
bull While the rollout of DCTT should initially employ an opt-in au-thorization approach the feasibility acceptability and value of opt-out approaches should continue to be evaluated As such opt-out approaches to consent should not be precluded by legislation
bull Congress should prohibit the commercial use of data collected for COVID-19 response by DCTT
bull Congress should prohibit discrimination on the basis of data col-lected by DCTT
bull If Congress is unable to enact suitable legislation state legislatures should work toward enacting similar laws for their jurisdictions A ldquomodelrdquo state law should be rapidly developed to facilitate na-tionwide uniformity of legal requirements
97
Public Health
Characteristics That Make Data Useful to
Public Health for Reducing Disease Transmission
bull Technologies or apps with the goal of enhancing public health ca-pacity to identify cases and trace contacts in order to control the spread of SARS-CoV-2 should be designed to match functionality with that goal
bull Technologies or apps may produce some false negatives or false positives but they should be accurate enough that public health authorities feel confident that they support and donrsquot detract from contact tracing efforts
bull DCTT approaches for public health should be designed to facili-tate the following
deg Identifying contacts including those who may not be easily found otherwise
deg Finding and notifying contacts rapidly before they develop symptoms if infected
deg Analyzing the nature of contact to determine whether con-tact is high medium or low risk to support decisions about whether quarantine should be mandatory should be volun-tary or is not needed
deg Following up with cases and contacts so that public health can provide resources to support isolation and quarantine at home
Recommendations
FIVE
98 Digital Contact Tracing for Pandemic Response
bull Data collected through DCTT should be made available to public health professionals and to researchers in de-identified form to support population-level epidemiologic analysis
Ethics
Collecting Data to Inform Policy and Practice
bull Reviews of DCTT systems must be conducted in part by an inde-pendent intermediary that has established the publicrsquos trust
bull Those who authorize use of DCTT within a particular jurisdic-tion or institution should continuously and systematically monitor the technologyrsquos performance in that context This should include monitoring for effectiveness and benefit monitoring for harms and monitoring for the fair distribution of both benefits and harms They should also monitor evidence that is being generated in other contexts about their selected technological solution and about other competing technologies
bull Data should be available to users that would permit them to further investigate their personal risk with public health officials or other health workers to add a layer of protection against unnecessary quarantine
Public Trust and Public Attitudes
bull More research into public attitudes is needed In particular in-depth qualitative research should examine public attitudes about perceptions of trust in DCTT among different communities which features of DCTT influence trust and the extent to which people are willing to provide different types of data through DCTT to help their community
bull States and localities that are considering adopting DCTT should engage with the public to increase their understanding of the acceptability of DCTT design features and uses among diverse communities
Recommendations 99
Designing Flexible Technology to Maximize
Public Health Utility While Respecting Other Values
bull Technology companies should not alone control the terms condi-tions or capabilities of DCTT nor should they presume to know what may be acceptable to members of the public
bull A ldquovalues in designrdquo approach to development of DCTT should be adopted (Flanagan Howe and Nissenbaum 2008 Knobel and Bowker 2011) Robust public- and user-engagement activities should be pursued to identify and incorporate to the extent pos-sible a range of values into the design of the technology These values may include privacy but also autonomy efficiency equity or others Technology design should reflect an appropriate balance and prioritization of identified values
bull Technology design should not be static but should be capable of evolving depending upon local conditions new evidence and changing preferences and priorities
bull DCTT should be designed to have a base set of features that pro-tect privacy with layers of additional capabilities that users may choose to activate An initial default should be that user location data are not shared but users should be provided with easy mech-anisms and prompts to allow for opting-in to this capability with encouragement to the public if and as it is shown to be critical to achieving public health goals
Policy Positions to Advance Widespread
Use of Digital Contact Tracing Technologies
bull DCTT use should not be mandated at this time given uncertainty about potential harms and benefits Additional technology user and real-world testing is needed
bull Incentives can be a useful complement to encouragements how-ever any incentives for users to install and use DCTT must be equitable should not be coercive and should align with effective use of the technology
100 Digital Contact Tracing for Pandemic Response
bull Trusted leaders should be enlisted to communicate effectively with the public about DCTT and encourage its use should the technol-ogy demonstrate some potential The limits of knowledge regard-ing effectiveness should also be explained along with what will be done to improve technological capabilities as understanding evolves
Disclosure and AuthorizationConsent
bull A clear and concise module consisting of basic disclosure and vol-untary authorization should be developed to accompany DCTT This module should not take the form of ldquoclickwraprdquo terms of service or end-user agreements but rather provide only essential information necessary for an individual to make a decision More detailed disclosures (such as FAQs in plain language) should be made easily accessible to those who wish to learn more with no hidden surprises
bull An opt-in approach to authorization should be instituted to ac-company initial DCTT rollout The feasibility and value of opt-out approaches should continue to be evaluated informed by what is technologically possible what local assessments of benefits and harms of the technology reveal over time and our evolving un-derstanding of the degree to which an opt-out approach is likely to increase or decrease utilization among different populations Opt-out approaches should not be precluded
Promoting Equity and Fairness
bull A commitment to equity means a commitment to ensuring that the benefits and burdens of DCTT are distributed fairly Public engagement is an important tool for assessing impact and to rectify inequities
bull States localities and institutions that recommend widespread use of DCTT should provide technology (eg mobile phones Blue-tooth devices) and free data packages to those who desire but lack access to these devices
Recommendations 101
bull If there are lower rates of adoption of DCTT systems in some identifiable communities public health authorities should iden-tify ways to compensate For example directing more non-DCTT resources and efforts toward those communities to meet specific needs that are elsewhere being supported by technology
bull If maps are generated based on DCTT to provide the public with the locations that COV+ individuals have visited steps must be taken to minimize the stigma and potential financial losses that could result from being identified as a hotspot
Instituting Transparent Governance and Oversight
bull Digital surveillance oversight committees should be established expeditiously with diverse and qualified membership to provide ethical and regulatory review prior to and concurrent with wide-spread use of a DCTT system
bull Understandable and publicly accessible rules must guide the col-lection access control use storage and combination of data by government authorities public and private institutions and other parties such as public health researchers
bull Only those data that are necessary and relevant for the public health response to COVID-19 should be collected and used
bull Identifiable data should be kept only for the period of time needed for the public health response to COVID-19
bull Identifiable data collected as part of this response should not be shared with anyone other than the relevant public health authorities without additional specific informed consent of individual users
bull Before a government or institution adopts a digital contact tracing program they should state the conditions under which the digital contact tracing program will be terminated
bull Future use of DCTT to advance public health or other efforts (eg use in seasonal flu surveillance) would require independent justification DCTT designed for public health use should not be used by law or immigration enforcement
102 Digital Contact Tracing for Pandemic Response
bull The principles offered in this guidance document apply both during and following the COVID-19 pandemic
Legislative
bull Congress should enact new legislation specifically tailored to fa-cilitate the use of DCTT as part of the public health response to COVID-19 while also protecting user privacy and ensuring data security
bull Congress should require DCTT developers to disclose to users in clear language the nature of the information that would be col-lected how it would be collected how it would be stored and for what purposes it may be used
bull While the rollout of DCTT should initially employ an opt-in au-thorization approach the feasibility acceptability and value of opt-out approaches should continue to be evaluated As such opt-out approaches to consent should not be precluded by legislation
bull Congress should prohibit the commercial use of data collected for COVID-19 response by DCTT
bull Congress should prohibit discrimination on the basis of data col-lected by DCTT
bull If Congress is unable to enact suitable legislation state legislatures should work toward enacting similar laws for their jurisdictions A ldquomodelrdquo state law should be rapidly developed to facilitate na-tionwide uniformity of legal requirements
Resources
103
US Government Response
White House
The White House and CDC 16 April 2020 ldquoOpening Up America Againrdquo The White House and the Centers for Disease Control and Prevention Available at httpswwwwhitehousegovopeningamerica
Congress
A Bill to Protect the Privacy of Consumersrsquo Personal Health Information
Proximity Data Device Data and Geolocation Data during the Corona-
virus Public Health Crisis S3663 116th Cong (2019-2020) Available at httpswwwcongressgovbill116th-congresssenate-bill3663
A Bill to Protect the Privacy of Health Information during a National
Health Emergency S3749 116th Cong (2019-2020) Available at httpswwwcongressgovbill116th-congresssenate-bill3749
To Protect the Privacy of Health Information during a National Health
Emergency HR6866 116th Cong (2019-2020) Available at httpswwwcongressgovbill116th-congresshouse-bill6866
Congressional Research Service
Foster Michael 16 April 2020 ldquoCOVID-19 Digital Surveillance and Privacy Fourth Amendment Considerationsrdquo Legal Sidebar LSB10449 Congres-sional Research Service httpscrsreportscongressgovproductpdfLSBLSB10449
Mulligan Stephen P and Chris D Linebaugh 25 March 2019 ldquoData Pro-tection Law An Overviewrdquo R45631 Congressional Research Service httpscrsreportscongressgovproductpdfRR45631
104 Resources
US Department of Health and Human Services (HHS) Centers for Disease Control and Prevention (CDC)
CDC 6 May 2020 ldquoInterim Guidance for Businesses and Employersrdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention Available at httpswwwcdcgovcoronavirus2019-ncov communityguidance-business-responsehtml
CDC 3 May 2020 ldquoDiscontinuation of Isolation for Persons with COVID-19 Not in Healthcare Settingsrdquo Interim Guidance Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention httpswwwcdc govcoronavirus2019-ncovhcpdisposition-in-home-patientshtml
CDC 30 April 2020 ldquoContact Tracingrdquo Get and Keep America Open Sup-porting States Tribes Localities and Territories Centers for Disease Control and Prevention httpswwwcdcgovcoronavirus2019-ncovphpopen-americacontact-tracinghtml
CDC 29 April 2020 ldquoInterim Guidelines for Collecting Handling and Testing Clinical Specimens from Persons for Coronavirus Disease 2019 (COVID-19)rdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention Available at httpswwwcdcgovcorona-virus2019-ncovlabguidelines-clinical-specimenshtml
CDC 29 April 2020 ldquoContact Tracing Part of a Multipronged Approach to Fight the COVID-19 Pandemicrdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention httpswwwcdcgov coronavirus2019-ncovphpprinciples-contact-tracinghtml
CDC 28 April 2020 ldquoPreliminary Criteria for the Evaluation of Digital Contact Tracing Tools for COVID-19rdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention Available at httpswwwcdc govcoronavirus2019-ncovdownloadsphpprelim-eval-criteria-digital- contact-tracingpdf
CDC 20 April 2020 ldquoDigital Contact Tracing Tools for COVID-19rdquo Corona- virus Disease 2019 (COVID-19) Centers for Disease Control and Preven-tion Available at httpswwwcdcgovcoronavirus2019-ncovdownloadsdigital-contact-tracingpdf
CDC 6 April 2020 ldquoOperational Considerations for the Identification of Healthcare Workers and Inpatients with Suspected COVID-19 in Non-US Healthcare Settingsrdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention Available at httpswwwcdcgovcoron virus2019-ncovhcpnon-us-settingsguidance-identify-hcw-patientshtml
HHS 2 April 2020 ldquoNotification of Enforcement Discretion under HIPAA to Allow Uses and Disclosures of Protected Health Information by Business
Resources 105
Associates for Public Health and Health Oversight Activities in Response to COVID-19rdquo 45 CFR Parts 160 and 164 Available at httpswwwhhsgovsitesdefaultfilesnotification-enforcement-discretion-hipaapdf
HHS 30 March 2020 Notification of Enforcement Discretion for Telehealth Remote Communications during the COVID-19 Nationwide Public Health Emergency US Department of Health and Human Services Office of Civil Rights Available at httpswwwhhsgovhipaafor-professionalsspecial- topicsemergency-preparednessnotification-enforcement-discretion- telehealthindexhtml
CDC 24 February 2020 ldquoLegal Authorities for Isolation and Quarantinerdquo Centers for Disease Control and Prevention Available at httpswww cdcgovquarantineaboutlawsregulationsquarantineisolationhtml
HHS 26 November 2019 ldquoPublic Health Emergency Declarationrdquo Public Health Emergency US Department of Health and Human Services Office of the Assistant Secretary for Preparedness and Response Available at httpswwwphegovPreparednesslegalPagesphedeclarationaspx
CDC September 2018 ldquoPublic Health Surveillance Preparing for the Futurerdquo Centers for Disease Control and Prevention Available at httpswww cdcgovsurveillancepdfsSurveillance-Series-Booklethpdf
CDC 2005 ldquoVI Plan for Surveillance of Contacts of SARS Case Supplement B SARS Surveillancerdquo Centers for Disease Control and Prevention Available at httpswwwcdcgovsarsguidanceb-surveillancecontactshtml
National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research 1979 ldquoThe Belmont Report Ethical Principles and Guidelines for the Protection of Human Subjects of Researchrdquo US Depart-ment of Health and Human Services Office of Human Research Protec-tions httpswwwhhsgovohrpregulations-and-policybelmont-reportread-the-belmont-reportindexhtml
US Equal Employment Opportunity Commission (EEOC)
EEOC 7 May 2020 ldquoWhat You Should Know About COVID-19 and the ADA the Rehabilitation Act and Other EEO Lawsrdquo US Equal Employ-ment Opportunity Commission Available at httpswwweeocgovwyskwhat-you-should-know-about-covid-19-and-ada-rehabilitation-act-and-other-eeo-laws
EEOC 21 March 2020 ldquoPandemic Preparedness in the Workplace and the Americans with Disabilities Actrdquo US Equal Employment Oppor-tunity Commission Available at httpswwweeocgovlawsguidancepandemic-preparedness-workplace-and-americans-disabilities-act
106 Resources
Federal Trade Commission (FTC)
FTC 2020 ldquoPrivacy amp Data Security Update 2019rdquo Federal Trade Commis-sion httpswwwftcgovsystemfilesdocumentsreportsprivacy-data- security-update-20192019-privacy-data-security-report-508pdf
FTC 2012 ldquoProtecting Consumer Privacy in an Era of Rapid Changerdquo FTC Report Recommendations for Businesses and Policymakers Fed-eral Trade Commission httpswwwftcgovsitesdefaultfilesdocumentsreportsfederal-trade-commission-report-protecting-consumer-privacy-era- rapid-change-recommendations120326privacyreportpdf
Occupational Safety and Health Administration (OSHA)
OSHA April 2020 ldquoGuidance on Preparing Workplaces for COVID-19rdquo OSHA 3990-03 2020 Department of Labor Occupational Safety and Health Act of 1970 Available at httpswwwoshagovPublicationsOSHA3990pdf
Customs and Border Protection (CBP)
CBP 4 January 2018 ldquoBorder Search of Electronic Mediardquo US Customs and Border Protection CDP DIRECTIVE NO 3340-049A Available at httpswwwcbpgovsitesdefaultfilesassetsdocuments2018-JanCBP-Directive-3340-049A-Border-Search-of-Electronic-Media-Compliantpdf
Other Governmental and Nongovernmental Organizations
Contact Tracing Surveillance PlansMethods
Simmons-Duffin Selena 7 May 2020 ldquoStates Nearly Doubled Plans for Contact Tracers since NPR Surveyed Them 10 Days Agordquo NPR Available at httpswwwnprorgsectionshealth-shots20200428846736937we-asked-all-50-states-about-their-contact-tracing-capacity-heres-what- we-learned
Resolve to Save Lives 2020 ldquoCOVID-19 Contact Tracing Playbookrdquo Vital Strategies Available at httpscontacttracingplaybookresolvetosavelives org
Simpson Erin and Adam Conner 22 April 2020 ldquoDigital Contact Tracing to Contain the Coronavirusrdquo Center for American Progress Avail-able at httpswwwamericanprogressorgissuestechnology-policynews20200422483521digital-contact-tracing-contain-coronavirus
Watson Crystal Anita Cicero James Blumenstock and Michael Fraser 10 April 2020 ldquoA National Plan to Enable Comprehensive COVID-19 Case Finding and Contact Tracing in the USrdquo Johns Hopkins Bloomberg School of Pub-
Resources 107
lic Health Center for Health Security and the Association of State and Ter-ritorial Health Officials Available at httpswwwcenterforhealthsecurity orgour-workpubs_archivepubs-pdfs2020200410-national-plan-to- contact-tracingpdf
Africa CDC 9 April 2020 ldquoGuidance on Contact Tracing for COVID-19 Pandemicrdquo Manuals Guidelines amp Frameworks African Union Africa CDC Available at httpsafricacdcorgdownloadguidance-on-contact- tracing-for-covid-19-pandemic
PIH 4 April 2020 ldquoPart I Testing Contact Tracing and Community Manage-ment of COVID-19rdquo PIH Guide | COVID-19 STOP COVID Partners in Health Available at httpswwwpihorgsitesdefaultfiles2020-04PIH_Guide_COVID_Part_I_Testing_Tracing_Community_Managment_4_4pdf
Hellewell Joel Sam Abbott Amy Gimma et al 28 February 2020 ldquoFea- sibility of Controlling COVID-19 Outbreaks by Isolation of Cases and Contactsrdquo The Lancet Global Health 8 e488ndash96 Available at https doiorg101016S2214-109X(20)30074-7
Contact Tracing EthicsPrinciplesGuidance
Center for Democracy amp Technology (CDT)ndashhttpscdtorginsightsCDT 30 April 2020 ldquoStatement of the Center for Democracy amp Technol-
ogy Regarding Use of Data to Fight COVID-19rdquo Center for Democ-racy amp Technology httpscdtorgwp-contentuploads202004 CDT-Statement-Regarding-Use-of-Data-to-Fight-COVID-19pdf
Shetty Ridhi 23 April 2020 ldquoData Use in the Fight against COVID-19 Should Treat People Equitably Not Exacerbate Long-Standing Dispar-itiesrdquo Center for Democracy amp Technology httpscdtorginsightsdata-use-in-the-fight-against-covid-19-should-treat-people-equitably-not-exacerbate-long-standing-disparities
Greenwood Dazza Gregory Nadeau Pagona Tsormpatzoudi Bryan Wilson Jeffrey Saviano and Alex ldquoSandyrdquo Pentland 30 April 2020 ldquoCOVID-19 Contact Tracing Privacy Principlesrdquo MIT Computational Law Report Available at httpslawmitedupubcovid19contacttracingprivacy- principlesrelease7
Editorial Board 29 April 2020 ldquoShow Evidence That Apps for COVID-19 Contact-Tracing Are Secure and Effectiverdquo Nature 580 (7805) 563ndash563 httpsdoiorg101038d41586-020-01264-1
ldquoJoint Statement on Contact Tracingrdquo 19 April 2020 Available at httpscryptobriefingcomwp-contentuploads202004Joint-State-ment-from-Researcherspdf
Hinch Robert Will Probert Anel Nurtay Michelle Kendall Chris Wymant
108 Resources
Matthew Hall Katrina Lythgoe et al 16 April 2020 ldquoEffective Configura-tions of a Digital Contact Tracing App A Report to NHSXrdquo Available at httpsgithubcomBDI-pathogenscovid-19_instant_tracingblobmasterReportndashEffective Configurations of a Digital Contact Tracing Apppdf
Kahn Gilmor Daniel 16 April 2020 ldquoPrinciples for Technology-Assisted Contact-Tracingrdquo White Paper American Civil Liberties Union httpswwwacluorgreportaclu-white-paper-principles-technology-assisted- contact-tracing
Carroll Anna and Samantha Stroman 16 April 2020 ldquoFind My Friends in a Pandemic The Future of Contact Tracing in Americardquo CSIS Commis-sion on Strengthening Americarsquos Health Security April 16 2020 httpshealthsecuritycsisorgarticlesfind-my-friends-in-a-pandemic-the-future- of-contact-tracing-in-america
Electronic Privacy Information Center Testimony to Congress 15 April 2020 ldquoEPIC to Congress Establish Privacy Safeguards for Digital Contact Tracingrdquo httpsepicorgtestimonycongressEPIC-HEC-Contact-Tracing-Apr2020pdf
Crocker Andrew Kurt Opsahl and Bennett Cyphers 10 April 2020 ldquoThe Chal-lenge of Proximity Apps for COVID-19 Contact Tracingrdquo Electronic Fron-tier Foundation April 10 2020 httpswwwefforgdeeplinks202004challenge-proximity-apps-covid-19-contact-tracing
Hart Vi Divya Siddarth Bethan Cantrell Lila Tretikov Peter Eckersley John Langford Scott Leibrand et al 3 April 2020 ldquoOutpacing the Virus Digital Response to Containing the Spread of COVID-19 While Mitigat-ing Privacy Risksrdquo Whitepaper 5 COVID-19 Rapid Response Impact Initiative Edmond J Safra Center for Ethics httpsdrivegooglecomfiled1vIN2AX-DDNW-S0aHq8xs0RJ2jkR_CckXview
Raskar Ramesh Isabel Schunemann Rachel Barbar Kristen Vilcans Jim Gray Praneeth Vepakomma Suraj Kapa Andrea Nuzzo Rajiv Gupta et al 19 March 2020 ldquoApps Gone Rogue Maintaining Personal Privacy in an Epidemicrdquo White Paper Private Kit MIT httpsarxivorgpdf2003 08567pdf
General (not COVID-19 specific) Statements of Principles Legal Frameworks Other Information
Nuffield Council on Bioethics 2020 ldquoGuide to the Ethics of Surveillance and Quarantine for Novel Coronavirusrdquo httpswwwnuffieldbioethicsorgassetspdfsGuide-to-the-ethics-of-surveillance-and-quarantine-for-novel-coronaviruspdf
Resources 109
Schwartz Adam 2020 ldquoHow EFF Evaluates Government Demands for New Surveillance Powersrdquo Electronic Frontier Foundation April 3 2020 httpswwwefforgdeeplinks202004how-eff-evaluates-government-demands- new-surveillance-powers
WHO 2 April 2020 ldquoCoronavirus Disease 2019 (COVID-19)rdquo Situation Report 73 World Health Organization httpswwwwhointdocsdefault-sourcecoronavirusesituation-reports20200402-sitrep-73-covid-19pdf
Nuffield Council on Bioethics 17 March 2020 ldquoEthical Considerations in Responding to the COVID-19 Pandemicrdquo Rapid Policy Briefing Nuffield Council on Bioethics httpswwwnuffieldbioethicsorgassetspdfs Ethical-considerations-in-responding-to-the-COVID-19-pandemicpdf
Nuffield Council on Bioethics 28 January 2020 ldquoResearch in Global Health Emergencies Ethical Issuesrdquo httpswwwnuffieldbioethicsorgassetspdfsRGHE_full_report1pdf
Vota Wayan 4 December 2019 ldquoEvery African Countryrsquos National eHealth Strategy or Digital Health Policyrdquo ICT Works httpswwwictworksorgafrican-national-ehealth-strategy-policy
ENISA 3 December 2019 ldquoPseudonymisation Techniques and Best Practicesrdquo European Union Agency for Cybersecurity httpswwwenisaeuropaeupublicationspseudonymisation-techniques-and-best-practices
ldquoMobile Cellular Subscriptions (per 100 People)rdquo 2018 World Telecommuni-cationICT Development Report International Telecommunication Union httpsdataworldbankorgindicatorITCELSETSP2
Levinson-Waldman Rachel 2018 ldquoCellphones Law Enforcement and the Right to Privacyrdquo Brennan Center for Justice New York University School of Law httpswwwbrennancenterorgsitesdefaultfiles2019-08Report_Cell_Surveillance_Privacypdf
Code of Federal Regulations Part 46ndashProtection of Human Subjects (45 CFR 46 US Department of Health and Human Services)
General Data Protection Regulation (REGULATION (EU) 2016679 (GDPR) European Union)
European Data Protection Supervisor ldquoNecessity and Proportionalityrdquo httpsedpseuropaeudata-protectionour-worksubjectsnecessity- proportionality_en
ISOIEC 38505-12017 Information technologymdashGovernance of ITmdashGover-nance of datamdashPart 1 Application of ISOIEC 38500 to the governance of data (2017 International Organization for Standardization (ISO))
WHO 2016 ldquoWHO Guidance for Managing Ethical Issues in Infectious Dis-ease Outbreaksrdquo World Health Organization httpswwwwhointethicspublicationsinfectious-disease-outbreaksen
110 Resources
CIOMS 2016 ldquoInternational Ethical Guidelines for Health-Related Research Involving HumansndashGuideline 22 Use of Data Obtained from the Online Environment and Digital Tools in Health Related Researchrdquo Council for International Organizations of Medical Sciences httpsciomschwp- contentuploads201701WEB-CIOMS-EthicalGuidelinespdf
Principles for Digital Development 2016 ldquoPrinciplesrdquo httpsdigitalprinciplesorgprinciples
GA4GH 2014 ldquoFramework for Responsible Sharing of Genomic and Health-Related Datardquo Global Alliance for Genomics amp Health 9 Decem-ber 2014 httpswwwga4ghorggenomic-data-toolkitregulatory-ethics- toolkitframework-for-responsible-sharing-of-genomic-and-health-related- data
WHO and PATH 2013 ldquoPlanning an Information Systems Project A Toolkit for Public Health Managersrdquo World Health Organization amp PATH httpspathazureedgenetmediadocumentsTS_opt_ict_toolkitpdf
ASTHO 2012 ldquoPublic Health Collection Use Sharing and Protection of Informationrdquo Issue Brief ASTHO Legal Preparedness Series Association of State and Territorial Health Officials httpswwwasthoorgProgramsPreparednessPublic-Health-Emergency-LawPublic-Health-and-Informat- ion-Sharing-ToolkitCollection-Use-Sharing-and-Protection-Issue-Brief
Cavoukian Ann 2010 ldquoPrivacy by Design The 7 Foundational Principlesrdquo Ontario Canada Information and Privacy Commissioner of Canada httpsiapporgmediapdfresource_centerpbd_implement_7found_ principlespdf
PHLS 2002 ldquoPrinciples of the Ethical Practice of Public Healthrdquo Public Health Leadership Society httpswwwaphaorg-mediafilespdfmembergroupsethicsethics_brochureashx
National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research 1979 ldquoThe Belmont Report Ethical Principles and Guidelines for the Protection of Human Subjects of Researchrdquo US Dept of Health amp Human Services Office of Human Research Protections httpswwwhhsgovohrpregulations-and-policybelmont-reportread-the-bel mont-reportindexhtml
Reopening ProposalsPlans (Broad)
United Kingdom Cabinet Office 2020 ldquoOur Plan to Rebuild The UK Govern-mentrsquos COVID-19 Recovery Strategyrdquo 12 May 2020 httpswwwgovukgovernmentpublicationsour-plan-to-rebuild-the-uk-governments-covid- 19-recovery-strategyour-plan-to-rebuild-the-uk-governments-covid-19- recovery-strategyfourteen-supporting-programmes
Resources 111
Shannon Joel Lorenzo Reyes and Doyle Rice 2020 ldquoAre Lockdowns Being Relaxed in My State Herersquos How America Is Reopening amid the Coronavirus Pandemicrdquo USA TODAY May 21 2020 httpswww usatodaycomstorynewshealth20200419coronavirus-lockdown- reopening-states-us-texas-florida5155269002
Romer Paul 2020 ldquoRoadmap to Responsibly Reopen Americardquo 23 April 2020 httpsroadmappaulromernetpaulromer-roadmap-reportpdf
Governor Larry Hogan 24 April 2020 ldquoMaryland Strong Roadmap to Recoveryrdquo httpsgovernormarylandgovwp-contentuploads202004MD_Strongpdf
NGA and ASTHO 21 April 2020 ldquoRoadmap to Recovery A Public Health Guide for Governorsrdquo National Governors Association and American State and Territorial Health Officials httpswwwngaorgwp-contentuploads202004NGA-Reportpdf
Allen Danielle Julius Krein Ganesh Sitaraman and E Glen Weyl 2020 ldquoNational Covid-19 Testing Action Planrdquo The Rockefeller Foundation httpswwwrockefellerfoundationorgwp-contentuploads202004TheRockefellerFoundation_WhitePaper_Covid19_4_22_2020pdf
Edmond J Safra Center for Ethics Harvard UniverityndashWhite PapersAllen Danielle Sharon Block Joshua Cohen et al 20 April 2020 ldquoRoad-
map to Pandemic Resiliencerdquo httpsethicsharvardedufilescenter- for-ethicsfilesroadmaptopandemicresilience_updated_42020_0pdf
Allen Danielle Lucas Stanczyk Rajiv Sethi Glen Weyl 25 March 2020 ldquoWhen Can We Go Outrdquo httpsdrivegooglecomfiled1gf21eYeNWwrR9OO5nzxn1jlv-RTmHkt0view
Mulheirn Ian Sam Alvis Lizzie Insall James Browne Christina Palmou 20 April 2020 ldquoA Sustainable Exit Strategy Managing Uncertainty Minimis-ing Harmrdquo Tony Blair Institute for Global Change httpsinstituteglobalsitesdefaultfilesinline-filesA20Sustainable20Exit20Strategy2C 20Managing20Uncertainty20Minimising20Harmpdf
Emanuel Zeke Neera Tanden Adam Conner Erin Simpson Nicole Rap-fogel and Maura Calsyn 2020 ldquoA National and State Plan to End the Coronavirus Crisisrdquo Center for American Progress April 3 2020 httpswwwamericanprogressorgissueshealthcarenews20200403482613national-state-plan-end-coronavirus-crisis
Gottlieb Scott Caitlin Rivers Mark McClellan Lauren Silvis and Crystal Watson 2020 ldquoNational Coronavirus Response A Road Map to Reopen-ingrdquo American Enterprise Institute httpswwwaeiorgresearch-productsreportnational-coronavirus-response-a-road-map-to-reopening
112 Resources
Digital Contact Tracing Experiences from Other Countries
Multiple
Woodhams Samuel 12 May 2020 ldquoCOVID-19 Digital Rights Trackerrdquo Top10VPN httpswwwtop10vpncomresearchinvestigationscovid-19- digital-rights-tracker
Fahim Kareem Min Joo Kim and Steve Hendrix 2 May 2020 ldquoCellphone Monitoring Is Spreading with the Coronavirus So Is an Uneasy Tolerance of Surveillancerdquo Washington Post (Washington DC) httpswww washingtonpostcomworldcellphone-monitoring-is-spreading-with-the- coronavirus-so-is-an-uneasy-tolerance-of-surveillance20200502 56f14466-7b55-11ea-a311-adb1344719a9_storyhtml
Ikram Umar Christer Mjaringset MD Anne-Marie Boxall Mylaine Breton Ines Gravey Holly Krelle Veacuteronique Raimond and Reginald D Williams II 30 April 2020 ldquoWhat Can the US Learn from Innovative Strategies Used in Other Countries to Respond to COVID-19rdquo The Commonwealth Fund httpswwwcommonwealthfundorgblog2020what-can-us-learn-innovative-strategies-used-other-countries-respond-covid-19
Jens-Henrik Jeppesen and Pasquale Esposito 29 April 2020 ldquoCOVID-19 European Data Collection and Contact Tracing Measuresrdquo Center for Democracy amp Technology httpscdtorginsightscovid-19-european- data-collection-and-contact-tracing-measures
Africa CDC 9 April 2020 ldquoGuidance on Contact Tracing for COVID-19 Pan-demicrdquo Manuals Guidelines amp Frameworks African Union Africa CDC httpsafricacdcorgdownloadguidance-on-contact-tracing-for-covid-19- pandemic
Heneghan Carl Jon Brassey and Tom Jefferson 6 April 2020 ldquoCOVID-19 What Proportion Are Asymptomaticrdquo Centre for Evidence-Based Medi-cine httpswwwcebmnetcovid-19covid-19-what-proportion-are- asymptomatic
Kharpal Arjun 30 March 2020 ldquoUse of Surveillance to Fight Coronavirus Raises Concerns about Government Power after Pandemic Endsrdquo CNBC httpswwwcnbccom20200327coronavirus-surveillance-used-by- governments-to-fight-pandemic-privacy-concernshtml
Canada
ldquoCommissioner Publishes Framework to Assess Privacy-Impactful Initiatives in Response to COVID19rdquo Office of the Privacy Commissioner of Canada 17 April 2020 httpswwwprivgccaenopc-newsnews-and-announcements 2020an_200417
Resources 113
China
Kraemer Moritz U G Chia-Hung Yang Bernardo Gutierrez Chieh-Hsi Wu Brennan Klein David M Pigott Open COVID-19 Data Working Group Louis du Plessis Nuno R Faria Ruoran Li William P Hanage John S Brownstein Maylis Layan Alessandro Vespignani Huaiyu Tian Chris-topher Dye Oliver G Pybus Samuel V Scarpino ldquoThe Effect of Human Mobility and Control Measures on the COVID-19 Epidemic in Chinardquo Science 368(6490) 493-497 DOI 101126scienceabb4218
Bi Qifang Yongsheng Wu Shujiang Mei Chenfei Ye Xuan Zou Zhen Zhang Xiaojian Liu Lan Wei Shaun A Truelove Tong Zhang Wei Gao Cong Cheng Xiujuan Tang Xiaoliang Wu Yu Wu Binbin Sun Suli Huang Yu Sun Juncen Zhang Ting Ma Justin Lessler and Teijian Feng ldquoEpidemiol-ogy and Transmission of COVID-19 in 391 Cases and 1286 of Their Close Contacts in Shenzhen China A Retrospective Cohort Studyrdquo The Lancet April 27 2020 DOIhttpsdoiorg101016S1473-3099(20)30287-5
Sun Kaiyuan and Ceacutecile Viboud ldquoImpact of contact tracing on SARS-CoV-2 transmissionrdquo The Lancet April 27 2020 DOIhttpsdoiorg101016S1473-3099(20)30357-1
Ferretti Luca Chris Wymant Michelle Kendall Lele Zhao Anel Nurtay Lucie Abeler-Doumlrner Michael Parker David Bonsall and Christophe Fraser 2020 ldquoQuantifying SARS-CoV-2 Transmission Suggests Epidemic Control with Digital Contact Tracingrdquo Science 368 (6491) httpsdoiorg101126scienceabb6936
Gan Nectar and David Culver 2020 ldquoChina Is Fighting the Coronavirus with a Digital QR code Herersquos How It Worksrdquo CNN Business April 16 2020 httpswwwcnncom20200415asiachina-coronavirus-qr-code-intl-hnkindexhtml
Mozur Paul Raymond Zhong and Aaron Krolik 2020 ldquoIn Coronavirus Fight China Gives Citizens a Color Code with Red Flagsrdquo New York Times (New York NY) March 1 2020 httpswwwnytimescom20200301businesschina-coronavirus-surveillancehtml
Germany
Schwartz Matthew S 2020 ldquoGermany Backs Away from Compiling Coro-navirus Contacts in a Central Databaserdquo NPR April 27 2020 https wwwnprorgsectionscoronavirus-live-updates20200427846046185 germany-backs-away-from-compiling-coronavirus-contacts-in-a-central- databaseutm_medium=RSSamputm_campaign=news
Busvine Douglas and Andreas Rinke 2020 ldquoGermany Flips to Apple-Google
114 Resources
Approach on Smartphone Contact Tracingrdquo Reuters April 26 2020 httpswwwreuterscomarticleus-health-coronavirus-europe-techgermany-flips-on-smartphone-contact-tracing-backs-apple-and-google-idUSKCN22807J
Hong Kong
ldquo lsquoStayHomeSafersquo Mobile App User Guiderdquo The Government of the Hong Kong
Special Administrative Region May 20 2020httpswwwcoronavirusgov hkengstay-home-safehtml
India
Dixit Pranav 2020 ldquoIndiarsquos Contact Tracing App Is All But Mandatory So This Programmer Hacked It So That He Always Appears Saferdquo BuzzFeed News May 12 2020 httpswwwbuzzfeednewscomarticlepranavdixitindia-aarogya-setu-hacked
Greenberg Andy 2020 ldquoIndiarsquos Covid-19 Contact Tracing App Could Leak Patient Locationsrdquo WIRED May 6 2020 httpswwwwiredcomstoryindia-covid-19-contract-tracing-app-patient-location-privacy
Alderson Elliot ldquoAarogya Setu The Story of a Failurerdquo Medium May 6 2020 httpsmediumcomfs0c131yaarogya-setu-the-story-of-a-failure- 3a190a18e34
OrsquoNeill Patrick Howell ldquoIndia Is Forcing People to Use Its COVID App Unlike Any Other Democracyrdquo MIT Technology Review May 6 2020 httpswwwtechnologyreviewcom202005071001360india-aarogya- setu-covid-app-mandatory
Israel
Hendrix Steve and Ruth Eglash 2020 ldquoIsrael Is Using Cellphone Surveil-lance to Warn Citizens You May Already Be Infectedrdquo Washington Post (Washington DC) March 19 2020 httpswwwwashingtonpostcomworldmiddle_eastisrael-is-using-cellphone-surveillance-to-warn-citizens-you-may-already-be-infected2020031968267294-69e7-11ea-b199-3a9799c54512_storyhtml
Singapore
ldquoTraceTogether Safer Togetherrdquo A Singapore Government Agency Website accessed May 21 2020 httpswwwtracetogethergovsg
Resources 115
South Korea
Korean Ministry of Health and Welfare ldquoConfirmed Patient Movement Path Websiterdquo Central Accident Remediation Headquarters accessed May 21 2020 httpncovmohwgokrbdBoardList_RealdobrdId=1ampbrdGu-bun=12ampncvContSeq=ampcontSeq=ampboard_id=ampgubun=
Kim Max S 2020 ldquoSeoulrsquos Radical Experiment in Digital Contact Tracingrdquo The New Yorker April 17 2020 httpswwwnewyorkercomnewsnews-deskseouls-radical-experiment-in-digital-contact-tracing
Watson Ivan and Sophie Jeong 2020 ldquoCoronavirus Mobile Apps Are Surging in Popularity in South Koreardquo CNN Business February 28 2020 httpseditioncnncom20200228techkorea-coronavirus-tracking-appsindexhtml
United Kingdom
Hern Alex and Kate Proctor 2020 ldquoUK May Ditch NHS Contact-Tracing App for Apple and Google Modelrdquo The Guardian May 7 2020 httpswwwtheguardiancomtechnology2020may07uk-may-ditch-nhs-contact- tracing-app-for-apple-and-google-model
Lovejoy Ben 2020 ldquoHands-on with UKrsquos NHS Contact Tracing App as the Test Goes Liverdquo 9-5 Mac Blog May 7 2020 https9to5maccom2020 0507nhs-contact-tracing
ldquoCoronavirus Test Track and Trace Plan Launched on Isle of Wightrdquo Depart-ment of Health and Social Care Press Release May 4 2020 httpswww govukgovernmentnewscoronavirus-test-track-and-trace-plan-launched- on-isle-of-wight
Leprince-Ringuet Daphne 2020 ldquoContact-Tracing Apps Why the NHS Said No to Apple and Googlersquos Planrdquo ZDNet April 28 2020 httpswww zdnetcomarticlecontact-tracing-apps-why-the-nhs-said-no-to-apple- and-googles-plan
Specific Digital ProductsApps
COVID-19
Vota Wayan 2020 Additional Proposed Coronavirus Solutions Google Doc-ument httpsdocsgooglecomspreadsheetsd15hkhdtGNzx7oHkO8Y 2MOiY83JsHjqxL4MhMGvlA_J6Ieditgid=579623365
Apple and Google nd ldquoPrivacy-Preserving Contact Tracingrdquo Apple Accessed May 19 2020 httpswwwapplecomcovid19contacttracing
116 Resources
Woodhams Samuel 12 May 2020 ldquoCOVID-19 Digital Rights Trackerrdquo Top10VPN httpswwwtop10vpncomresearchinvestigationscovid-19-digital-rights-tracker
Starobinski David and Johannes Becker 30 April 2020 ldquoHow Apple and Goo-gle Will Let Your Phone Warn You If Yoursquove Been Exposed to the Corona-virusrdquo The Conversation Accessed May 21 2020 httptheconversation comhow-apple-and-google-will-let-your-phone-warn-you-if-youve-been-exposed-to-the-coronavirus-136597
Morrison Sara 2020 ldquoThe Apple-Google Contact Tracing Tool Gets a Beta Release and a New Risk Level Featurerdquo Vox April 24 2020 httpswwwvoxcomrecode202042421234420apple-google-contact-tracing- exposure-notification-update
Commonwealth Centre for Digital Health 9 April 2020 ldquo[Webinar] CWCDH Digital Response to COVID-19rdquo 5040 httpswwwyoutubecomwatchv=ZyE_KRWLtC8ampfeature=youtube
Johns Hopkins Medicine 23 April 2020 ldquoJohns Hopkins Medicine Remote Monitoring Program for Faculty Staff and Providers Exposed to COVID-19rdquo Johns Hopkins Medicine Occupational Health Services httpswww hopkinsmedicineorghsecovid19_emocha
MIT nd ldquoPrivate Kit Safe Paths Privacy-by-Design Covid19 Solutions Using GPS+Bluetooth for Citizens and Public Health Officialsrdquo Safepaths Accessed May 20 2020 httpssafepathsmitedu
Related Disease Detection Projects
ldquoCOVID Controlrdquo Google Play accessed May 21 2020 httpsplaygoogle comstoreappsdetailsid=jhueduJohnsHopkinsCOVIDControl
ldquoCOVID Symptom Trackerrdquo Created by Massachusetts General Hospital the Harvard TH Chan School of Public Health Kingrsquos College London and Stanford University School of Medicine accessed May 21 2020 httpscovidjoinzoecomus
ldquoSee How Your Community Is Moving around Differently Due to COVID-19rdquo Google COVID-19 Mobility Reports accessed May 21 2020 httpswww googlecomcovid19mobility
Drew David A Long H Nguyen Claire J Steves Cristina Menni Maxim Freydin Thomas Varsavsky Carole H Sudre M Jorge Cardoso Sebastien Ourselin Jonathan Wolf Tim D Spector Andrew T Chan and COPE Consortium 2020 ldquoRapid Implementation of Mobile Technology for Real-Time Epidemiology of COVID-19rdquo Science 05 May 2020 DOI 101126scienceabc0473
Resources 117
Tress Luke 2020 ldquoMaccabi Medial EarlySign Develop Algorithm to Identify High-Risk COVID-19 Casesrdquo Times of Israel April 22 2020 httpswwwtimesofisraelcommaccabi-medial-earlysign-develop-algo rithm-to-identify-high-risk-covid-19-cases
ldquoPCR Diagnostic Testing for SARS-CoV-2rdquo Center for Health Security Johns Hopkins Bloomberg School of Public Health last modified April 17 2020 httpswwwcenterforhealthsecurityorgresourcesCOVID-19COVID-19-fact-sheets200130-nCoV-diagnostics-factsheetpdf
Radin Jennifer M Nathan E Wineinger Eric J Topol and Steven R Steinhubl 2020 ldquoHarnessing Wearable Device Data to Improve State-Level Real-Time Surveillance of Influenza-Like Illness in the USA A Population-Based Studyrdquo The Lancet January 16 2020 DOIhttpsdoiorg101016S2589-7500(19)30222-5
Polling
Hargittai Eszter Minh Hao Nguyen Jaelle Fuchs Jonathan Gruber Will Marler Amanda Hunsaker and Goumlkccedile Karaoglu 2020 ldquoCovid-19 Study on Digital Media and the Coronavirus Pandemicrdquo Internet Use and Society Division Institute of Communication and Media Research University of Zurich httpwebuseorgcovid
Navigator Research 2020 ldquoPublic Opinion on Coronavirus Navigator Updaterdquo Navigating Coronavirus (blog) May 21 2020 https navigatorresearchorgnavigating-coronavirus
Russonello Giovanni 2020 ldquoBig Government For Now Most Americans Say Bring It Onrdquo The New York Times May 1 2020 sec US httpswww nytimescom20200501uspoliticscoronavirus-spending-pollshtml
ldquoWashington Post-University of Maryland National Pollrdquo 2020 Washington Post April 21 2020 httpswwwwashingtonpostcomcontext washington-post-university-of-maryland-national-poll-april-21-26-20203583b4e9-66be-4ed6-a457-f6630a550ddf
Kirzinger Ashley Liz Hamel Cailey Muntildeana Audrey Kearney and Mollyann Brodie 2020 ldquoKFF Health Tracking Poll ndash Late April 2020 Coronavirus Social Distancing and Contact Tracingrdquo Kaiser Family Foundation (blog) April 24 2020 httpswwwkfforgcoronavirus-covid-19issue-briefkff-health-tracking-poll-late-april-2020
NYC DOH 2020 ldquoAge-Adjusted Rates of Lab Confirmed COVID-19 Non-hospitalized Cases Estimated Non-Fatal Hospitalized Cases and Persons Known to Have Died per 100000 by RaceEthnicity Grouprdquo New York
118 Resources
City Department of Health httpswww1nycgovassetsdohdownloadspdfimmcovid-19-deaths-race-ethnicity-04242020-1pdf
Anderson Monica and Brooke Auxier 2020 ldquoMost Americans Donrsquot Think Cellphone Tracking Will Help Limit COVID-19 Are Divided on Whether Itrsquos Acceptablerdquo Pew Research Center (blog) April 16 2020 httpswwwpewresearchorgfact-tank20200416most-americans-dont-think-cellphone-tracking-will-help-limit-covid-19-are-divided-on-whether-its-acceptable
Daly Kyle 2020 ldquoExclusive Americans Wary of Giving up Data to Fight Coronavirusrdquo Axios April 3 2020 httpswwwaxioscomexclusive- americans-wary-of-giving-up-data-to-fight-coronavirus-330fc1d9-8b99-4a51-871b-25ee0e0591f2html
Elliott Douglas J Ana Kreacic Lorenzo Milans del Bosch and Lisa Quest nd ldquoData-Sharing in the Time of Coronavirusrdquo Oliver Wyman Forum Accessed May 19 2020 httpswwwoliverwymanforumcomfuture-of-data2020aprdata-sharing-in-the-time-of-coronavirushtml
Auxier Brooke Lee Rainie Monica Anderson Andrew Perrin Madhu Kumar and Erica Turner 2019 ldquoAmericans and Privacy Concerned Confused and Feeling Lack of Control Over Their Personal Informationrdquo Pew Research
Center Internet Science amp Tech (blog) November 15 2019 httpswww pewresearchorginternet20191115americans-and-privacy-concerned- confused-and-feeling-lack-of-control-over-their-personal-information
EPIC 2020 ldquoPublic Opinion on Privacyrdquo Electronic Privacy Information Center January 22 2020 httpsepicorgprivacysurvey
Pew Research Center 2020 ldquoDemographics of Mobile Device Ownership and Adoption in the United Statesrdquo Internet amp Technology Pew Research Cen-ter httpswwwpewresearchorginternetfact-sheetmobile
Rodrigues Rafaela Alina Husain Amanda Couture-Carron Leslye E Orloff and Nawal H Ammar 2018 ldquoPromoting Access to Justice for Immigrant and Limited English Proficient Crime Victims in an Age of Increased Immi-gration Enforcement Initial Report from a 2017 National Surveyrdquo Wash-ington DC National Immigrant Womenrsquos Advocacy Project American University Washington College of Law httpniwaplibrarywclamericaneduwp-contentuploadsImmigrant-Access-to-Justice-National-Reportpdf
Center for Survey Measurement (CSM) MEMORANDUM FOR Associate Directorate for Research and Methodology (ADRM) 2017 ldquoRespondent Confidentiality Concernsrdquo September 20 2017 httpswww2censusgovcacnacmeetings2017-11Memo-Regarding-Respondent-Confidentiality- Concernspdf
Resources 119
Pew Research Center 2017 ldquoUS Muslims Concerned About Their Place in Society but Continue to Believe in the American Dreamrdquo Religion amp Public Life Pew Research Center httpswwwpewforumorg20170726findings-from-pew-research-centers-2017-survey-of-us-muslims
Popular Press
Dixit Pranav 2020 ldquoIndiarsquos Contact Tracing App Is All But Mandatory So This Programmer Hacked It So That He Always Appears Saferdquo BuzzFeed News May 12 2020
Mills Rodrigo Chris 2020 ldquoDigital Contact Tracing Is Becoming Available but Is It Effectiverdquo The Hill May 7 2020 httpsthehillcompolicytechnology 496498-digital-contact-tracing-is-becoming-available-but-is-it-effective
Ingram David 2020 ldquoApple Google Push Makers of Coronavirus Apps Not to Record User Locationrdquo May 4 2020 NBC News Accessed May 19 2020 httpswwwnbcnewscomtechtech-newscoronavirus-apps-won-t-be-able- record-users-location-apple-n1199586
McMinn Sean 2020 ldquoMobile Phone Data Show More Americans Are Leav- ing Their Homes Despite Ordersrdquo NPR May 1 2020 httpswwwnpr org20200501849161820mobile-phone-data-show-more-americans- are-leaving-their-homes-despite-orders
Valentino-DeVries Jennifer Natasha Singer and Aaron Krolik 2020 ldquoA Scram-ble for Virus Apps That Do No Harmrdquo The New York Times April 29 2020 sec Business httpswwwnytimescom20200429 businesscoronavirus-cellphone-apps-contact-tracinghtml
Giglio Mike 2020 ldquoWould You Sacrifice Your Privacy to Get Out of Quaran-tinerdquo The Atlantic April 22 2020 httpswwwtheatlanticcompoliticsarchive202004coronavirus-pandemic-privacy-civil-liberties-911609172
Bradshaw Tim 2020 ldquo2 Billion Phones Cannot Use Google and Apple Con-tact-Tracing Tech | Ars Technicardquo Ars Technica April 20 2020 httpsarstechnicacomtech-policy2020042-billion-phones-cannot-use-google-and-apple-contract-tracing-tech
Reston Maeve Kristina Sgueglia and Cheri Mossburg 2020 ldquoGovernors on East and West Coasts Form Pacts to Decide When to Reopen Economiesrdquo CNN Politics April 13 2020 httpswwwcnncom20200413politicsstates-band-together-reopening-plansindexhtml
Cochrane Emily Claire Cain Miller and Jim Tankersley 2020 ldquoTrump Admin-istration Scales Back Paid Leave in Coronavirus Relief Lawrdquo The New York Times April 2 2020 sec US httpswwwnytimescom20200402uspoliticscoronavirus-paid-leavehtml
120 Resources
Kim Nemo 2020 ldquolsquoMore Scary than Coronavirusrsquo South Korearsquos Health Alerts Expose Private Livesrdquo The Guardian March 6 2020 sec World news httpswwwtheguardiancomworld2020mar06more-scary- than-coronavirus-south-koreas-health-alerts-expose-private-lives
Hickey Matt 2014 ldquoCarriers Can Now Install Apps On Android Handsets Without Customersrsquo Permissionrdquo Forbes December 1 2014 httpswww forbescomsitesmatthickey20141201carriers-can-now-install-apps-on- android-handsets-without-customers-permission
Commentaries
Giubilini Alberto 2020 ldquoContact-Tracing Apps and the Future COVID-19 Vaccination Should Be Compulsory Social Technological and Pharmaco-logical Immunisationrdquo Practical Ethics (blog) May 6 2020 httpblo gpracticalethicsoxacuk202005contact-tracing-apps-and-the-future- covid-19-vaccination-should-be-compulsory-social-technological-and- pharmacological-immunisation
Landau Susan Christy Lopez and Laura Moy 2020 ldquoThe Importance of Equity in Contact Tracingrdquo Lawfare (blog) May 1 2020 httpswww lawfareblogcomimportance-equity-contact-tracing
Schaefer G Owen and Angela Ballantyne 2020 ldquoDownloading COVID-19 Contact Tracing Apps Is a Moral Obligationrdquo Journal of Medical Ethics
Blog (blog) May 4 2020 httpsblogsbmjcommedical-ethics20200504downloading-covid-19-contact-tracing-apps-is-a-moral-obligation
OrsquoNeill Patrick Howell 2020 ldquoFive Things We Need to Do to Make Contact Tracing Really Workrdquo MIT Technology Review April 28 2020 httpswwwtechnologyreviewcom202004281000714five-things-to-make- contact-tracing-work-covid-pandemic-apple-google
Doffman Zak 2020 ldquoCOVID-19 Contact Tracing Why Apple And Google Canrsquot Make This Workrdquo Forbes April 27 2020 httpswwwforbescomsiteszakdoffman20200427this-is-the-contact-tracing-worry-even-apple- and-google-cant-resolve
All Tech is Human 2020 The Ethics of Contact Tracing for COVID-19 httpswwwyoutubecomwatchv=59mKUAVDhdkampt=626s
Canca Cansu 2020 ldquoWhy lsquoMandatory Privacy-Preserving Digital Contact Tracingrsquo Is the Ethical Measure against COVID-19rdquo Medium May 4 2020 httpsmediumcomcansucancawhy-mandatory-privacy-preserv-ing-digital-contact-tracing-is-the-ethical-measure-against-covid-19-a0d-143b7c3b6
Goodman Bryce 2020 ldquoCOVID and Contact Tracing When Social Justice
Resources 121
Demands Mass Surveillancerdquo Medium April 10 2020 httpsmediumcombwgoodmancovid-and-contact-tracing-when-social-justice-demands-mass-surveillance-18d419b8cc5
Gray Rosie and Caroline Haskins 2020 ldquoThey Were Opposed To Govern-ment Surveillance Then The Coronavirus Pandemic Beganrdquo BuzzFeed News March 30 2020 httpswwwbuzzfeednewscomarticlerosiegraythey-were-opposed-to-government-surveillance-then-the
Cegłowski Maciej 2020 ldquoWe Need A Massive Surveillance Programrdquo Idle
Words (blog) March 23 2020 httpsidlewordscom202003we_need_ a_massive_surveillance_programhtm
Academic Literature
COVID-19 Specific
Abeler J Baumlcker M Buermeyer U Zillessen H (2020) COVID-19 Contact Tracing and Data Protection Can Go Together JMIR mHealth and uHealth 8(4) e19359 doi 10219619359
Altmann S Milsom L Zillessen H et al (2020) Acceptability of App-Based Contact Tracing for COVID-19 Cross-Country Survey Evidence Preprint
Berke A Bakker M Vepakomma P Larson K Pentland A (2020) Assessing Disease Exposure Risk with Location Data A Proposal for Cryptographic Preservation of Privacy arXiv arXiv200314412ndashMarch 2020
Bonsail D Parker M Fraser C (2020) Sustainable Containment of COVID-19 Using Smartphones in China Scientific and Ethical Underpinnings for Implementation of Similar Approaches in Other Settings Unpublished working paper
Bradshaw WJ Alley EC Huggins JH Lloyd AL Esvelt KM (2020) Bidirectional Contact Tracing Is Required for Reliable COVID-19 Control Preprint via MedRxiv
Braithwaite I Callender T Bullock M Aldridge R (2020) Automated and Semi-Automated Contact Tracing Protocol for a Rapid Review of Avail-able Evidence and Current Challenges to Inform the Control of COVID-19 Preprint via medRxiv doi httpsdoiorg1011012020041420063636
Bulchandani Bannerjee V Shivam S Moudgalya S Sondhi SL (2020) Digital Herd Immunity and COVID-19 Preprint via medRxiv doi httpsdoiorg1011012020041520066720
Cheng H Jian S Liu D (2020) Contact Tracing Assessment of COVID-19 Transmission Dynamics in Taiwan and Risk at Different Exposure Periods Before and After Symptom Onset JAMA Internal Medicine doi101001jamainternmed20202020
122 Resources
Cho H Ippolito D Yu YW (2020) Contact Tracing Mobile Apps for COVID-19 Privacy Considerations and Related Trade-offs httpsarxivorgpdf200311511pdf
Devakumar D Geordan S Bhopal SS Abubakar I (2020) Racism and dis-crimination in COVID-19 responses The Lancet 395(10231) 1194 doi 101016S0140-6736(20)30792-3
Drew D Nguyen L Steves C et al (2020) Rapid Implementation of Mobile Technology for Real-Time Epidemiology of COVID-19 Science published online May 5 2020
Ferretti L Wymant C Kendall M et al (2020) Quantifying SARS-CoV-2 Trans-mission Suggests Epidemic Control with Digital Contact Tracing Science doi 101126scienceabb6936
Fraser C Abeler-Doumlrner L Ferretti L et al (2020) Digital Contact Tracing Comparing the Capabilities of Centralised and Decentralised Data Archi-tectures to Effectively Suppress the COVID-19 Epidemic While Maximizing Freedom of Movement and Maintaining Privacy Preprint
Leith DJ Farrell S (2020) Coronavirus Contact Tracing Evaluating The Poten-tial Of Using Bluetooth Received Signal Strength For Proximity Detection Preprint
Jayant Limaye R Sauer M Ali J et al (2020) Building Trust While Influencing Online COVID-19 Content iIn the Social Media World The Lancet Digital Health
Mahmood S Hasan K Colder Carras M Labrique A (2020) Global Prepared-ness Aagainst COVID-19 We Must Leverage the Power of Digital Health JMIR Public Health Surveill 20206(2)e18980 DOI 10219618980
Mello M Wang CJ (2020) Ethics and Governance for Digital Disease Surveil-lance Science 11 May 2020 eabb9045 DOI 101126scienceabb9045
Park S Jeehyun Choi G Ko H (2020) Information TechnologyndashBased Tracing Strategy in Response to COVID-19 in South KoreamdashPrivacy Controversies JAMA doi101001jama20206602
Parker M Fraser C Abeler-Doumlrner L Bonsall D (2020) Ethics of Instantaneous Contract Tracing Using Mobile Phone Apps in the Control of the COVID-19 Pandemic Journal of Medical Ethics Published Online May 4 2020
Ethics and Digital Disease Detection
Aiello A Renson A Civich P (2020) Social Mediandash and Internet-Based Dis-ease Surveillance for Public Health Annual Review of Public Health 41 101ndash118 doi 101146annurev-publhealth-040119-094402
Ali J DiStefano M Coates McCall I et al (2019) Ethics of Mobile Phone Sur-
Resources 123
veys to Monitor Non-Communicable Disease Risk Factors in Low- and Middle-Income Countries A Global Stakeholder Survey Global Public Health 14(8) 1167ndash1181
Ali J Labrique A Gionfriddo K et al (2017) Ethics Considerations in Global Mobile Phone-Based Surveys of Noncommunicable Diseases A Concep-tual Exploration Journal of Medical Internet Research 19(5) e110 doi 102196jmir7326
Brownstein J Freifeld C Madoff L (2009) Digital Disease DetectionndashHar-nessing the Web for Public Health Surveillance New England Journal of Medicine 360(21) 2153ndash2157 doi 101056NEJMp0900702
Danquah LO Hasham N MacFarlane M et al (2019) Use of a Mobile Application for Ebola Contact Tracing and Monitoring in Northern Sierra Leone A Proof-of-Concept Study BMC Infectious Diseases 19 810
Degeling C Carter S van Oijen A et al (2020) Community Perspectives on the Benefits and Risks of Technologically Enhanced Communicable Disease Surveillance Systems A Report on Four Community Juries BMC Medical Ethics 21 31 doi 101186s12910-020-00474-6
DeJong B Badou G Luten J et al (2019) Ethical Considerations for Movement Mapping to Identify Disease Transmission Hotspots Emerging Infectious Diseases 25(7) e181421 doi 103201eid2507181421
Denecke K (2017) An Ethical Assessment Model for Digital Disease Detec-tion Technologies Life Sciences Society and Policy 13 16 doi 101186s40504-017-0062-x
Genevieve LD Martani A Wangmo T et al (2019) Participatory Disease Surveillance Systems Ethical Framework Journal of Medical Internet Research 21(5) e12273 doi10219612273
Gilbert G Degeling C Johnson J (2017) Communicable Disease Surveillance Ethics in the Age of Big Data and New Technology Asian Bioethics Review 11 173-187 doi 101007s41649-019-00087-1
Iwaya LH Li J Fischer-Hubner S et al (2019) E-Consent for Data Privacy Consent Management for Mobile Health Technologies in Public Health Surveys and Disease Surveillance Studies in health technology and infor-matics 264 1223-1227 doi 103233SHTI190421
Kostkova P (2018) Disease Surveillance Data Sharing for Public Health The Next Ethical Frontiers Life Sciences Society and Policy 14 16 doi 101186s40504-018-0078-x
Mahmood S Hasan K Colder Carras M Labrique A (2020) Global Prepared-ness against COVID-19 We Must Leverage the Power of Digital Health JMIR Public Health Surveillance 6(2)e18980
124 Resources
Smolinski MS Crawley AW Baltrusaitis K et al (2015) Flu Near You Crowd-sourced Symptom Reporting Spanning 2 Influenza Seasons AJPH 105(10) 2124ndash2130
Wojcik O Brownstein J Chunara R Johansson M (2014) Public Health for the People Participatory Infectious Disease Surveillance in the Digital Age Emerging Themes in Epidemiology 11 7 doi 1011861742-7622-11-7
Other Works
Bernstein J Holroyd TA Atwell JE et al (2019) Rockland Countyrsquos Proposed Ban against Unvaccinated Minors Balancing Disease Control Trust and Liberty Vaccine 37(30) 3933ndash3935
Berry SM Petzold EA Dull P et al (2017) A Response Adaptive Randomiza-tion Platform Trial for Efficient Evaluation of Ebola Virus Treatments A Model for Pandemic Response Clinical Trials 13(1) 22ndash30 doi 101177 1740774515621721
Beukenhorst AL Schultz DM McBeth J (2017) Using Smartphones for Research outside Clinical Settings How Operating Systems App Develop-ers and Users Determine Geolocation Data Quality in mHealth Studies MEDINFO 2017 Precision Healthcare through Informatics
Bourne P (2015) Confronting the Ethical Challenges of Big Data in Public Health Plos Computational Biology 11(2) e1004073 doi 101371journal pcbi1004073
Doerr M Suver C Wilbanks J (2016) Developing a Transparent Par ticipant-Navigated Electronic Informed Consent for Mobile-Mediated Research (April 22 2016) Available at SSRN httpsssrncomabstract=2769129 or httpdxdoiorg102139ssrn2769129
Dredze M Paul MJ Bergsma S Tran H (2013) Carmen A Twitter Geolocation System with Applications to Public Health Expanding the Boundaries of Health Informatics Using Artificial Intelligence Papers from the AAAI 2013 Workshop
Eckhoff PA Tatem AJ (2015) Digital methods in epidemiology can transform disease controlInternational Health Volume 7 Issue 2 March 2015 Pages 77ndash78 httpsdoiorg101093inthealthihv013
Edelstein M Lee L Herten-Crabb A Heymann D Harper D (2018) Strength-ening Global Public Health Surveillance through Data and Benefit Sharing Emerging Infectious Diseases 24(7) 1324ndash1330 doi 103201eid2407151830
Faden R Beauchamp T (1986) A History and Theory of Informed Consent Oxford University Press ISBN 9780199748655
Resources 125
Fairchild A Bayer R (2004) Ethics and the Conduct of Public Health Surveil-lance Science 303(5658) 631ndash632
Flanagan M Howe DC Nissenbaum H (2008) Embodying Values in Technol-ogy Theory and Practice In Information Technology and Moral Philoso-phy van den Hoven J amp Weckert J (eds) Cambridge Cambridge Univer-sity Press 322ndash353
Fraccaro P Beukenhorst A Sperrin M et al (2019) Digital Biomarkers from Geolocation Data in Bipolar Disorder and Schizophrenia A Systematic Review Journal of the American Medical Informatics Association Volume 26 Issue 11 November 2019 Pages 1412ndash1420
Furlanello C Merler S Menegon S et al (2002) New WEBGIS Technologies for Geo-location of Epidemiological Data An Application for the Sur-veillance of the Risk of Lyme borreliosis Disease In Giornale Italiano di Aritmologia e Cardiostimolazione Special Issue Proceedings of the 15th Int Congress ldquoThe New Frontiers of Arrhythmiasrdquo v 5 n1 Mar 2002 241ndash245
Gibson DG Wosu AC Pariyo GW et al (2019) Effect of Airtime Incentives on Response and Cooperation Rates in Non-communicable Disease Interactive Voice Response Surveys Randomised Controlled Trials in Bangladesh and Uganda BMJ Global Health 4(5)
Knobel C Bowker GC (2011) Computing Ethics Values in Design Communi-cations of the acm 54(7) 26ndash28
Lee R Cui RR Muessig KE et al (2015) Incentivizing HIVSTI Testing A Sys-tematic Review of the Literature AIDS and Behavior 18(5) 905ndash912 doi 101007s10461-013-0588-8
Lorchan LT Wyatt J (2014) mHealth and Mobile Medical Apps A Frame-work to Assess Risk and Promote Safer Use Journal of Medical Internet Research 16(9) e210 doi 102196jmir3133
Mathews S McShea M Hanley C et al (2019) Digital Health A Path to Vali-dation Digital Medicine 238 doi 101038s41746-019-0111-3
Moore S Tasse A Thorogood A et al (2017) Consent Processes for Mobile App Mediated Research Systematic Review JMIR mHealth amp uHealth 5(8) e126 DOI 102196mhealth7014
Pallman P Bedding AW Choodari-Oskooei B et al (2018) Adaptive Designs in Clinical Trials Why Use Them and How to Run and Report Them BMC Medicine 16(29)
Rennie S Buchbinder M Juengst E et al (2020) Scraping the Web for Public Health Gains Ethical Considerations from a lsquoBig Datarsquo Research Project on HIV and Incarceration Public Health Ethics doi 101093phephaa006
126 Resources
Rithalia A McDaid C Suekarran S (2009) Impact of Presumed Consent for Organ Donation on Donation Rates A Systematic Review BMJ 2009 338
Singer E Ye C (2012) The Use and Effects of Incentives in Surveys The ANNALS of the American Academy of Political and Social Science 645(1) 112ndash141
Vayena E Blassime A (2018) Health Research with Big Data Time for Systemic Oversight J Law Med Ethics 2018 Mar 46(1) 119ndash129
Vayena E Mastroianni A Kahn J (2012) Ethical Issues in Health Research with Novel Online Sources American Journal of Public Health 102(12) 2225ndash2230 doi 102105AJPH2012300813
Vayena E Salatheacute M Madoff L Brownstein J (2015) Ethical Challenges of Big Data in Public Health Plos Computational Biology 11(2) e1003904 doi 101371journalpcbi1003904
Xafis V Schaefer GO Labude MK et al (2019) An Ethics Framework for Big Data in Health and Research Asian Bioethics Review volume 11 pages 227ndash254(2019)
Works Cited
127
Africa CDC 2020 ldquoGuidance on Contact Tracing for COVID-19 Pandemicrdquo Manuals Guidelines amp Frameworks African Union Africa CDC httpsafricacdcorgdownloadguidance-on-contact-tracing-for-covid-19- pandemic
Aiello Allison E Audrey Renson and Paul N Zivich 2020 ldquoSocial Mediandash and Internet-Based Disease Surveillance for Public Healthrdquo Annual Review of Public Health 41 (1) 101ndash18 httpsdoiorg101146annurev-publhealth-040119-094402
Ali Joseph Michael J DiStefano Iris Coates McCall Dustin G Gibson Gulam Muhammed Al Kibria George W Pariyo Alain B Labrique and Adnan A Hyder 2019 ldquoEthics of Mobile Phone Surveys to Monitor Non-Communicable Disease Risk Factors in Low- and Middle-Income Countries A Global Stakeholder Surveyrdquo Global Public Health 14 (8) 1167ndash81 httpsdoiorg1010801744169220191566482
Ali Joseph Alain B Labrique Kara Gionfriddo George Pariyo Dustin G Gibson Bridget Pratt Molly Deutsch-Feldman and Adnan A Hyder 2017 ldquoEthics Considerations in Global Mobile Phone-Based Surveys of Noncommunicable Diseases A Conceptual Explorationrdquo Journal of Medi-cal Internet Research 19 (5) e110 httpsdoiorg102196jmir7326
Allen Danielle Julius Krein Ganesh Sitaraman and E Glen Weyl 2020 ldquoNational Covid-19 Testing Action Planrdquo The Rockefeller Foundation httpswwwrockefellerfoundationorgwp-contentuploads202004TheRockefellerFoundation_WhitePaper_Covid19_4_22_2020pdf
Altmann Samuel Luke Milsom Hannah Zillessen Raffaele Blasone Frederic Gerdon Ruben Bach Frauke Kreuter Daniele Nosenzo Severine Tous-saert and Johannes Abeler 2020 ldquoAcceptability of App-Based Contact Tracing for COVID-19 Cross-Country Survey Evidencerdquo MedRxiv May 2020050520091587 httpsdoiorg1011012020050520091587
Anderson Monica and Brooke Auxier 2020 ldquoMost Americans Donrsquot Think Cellphone Tracking Will Help Limit COVID-19 Are Divided on Whether
128 Works Cited
Itrsquos Acceptablerdquo Pew Research Center (blog) April 16 2020 https wwwpewresearchorgfact-tank20200416most-americans-dont-think-cellphone-tracking-will-help-limit-covid-19-are-divided-on-whether-its- acceptable
Anderson Monica and Andrew Perrin 2017 ldquoDisabled Americans Less Likely to Use Technologyrdquo Pew Research Center (blog) April 7 2017 httpswwwpewresearchorgfact-tank20170407disabled-americans- are-less-likely-to-use-technology
Apple and Google nd ldquoPrivacy-Preserving Contact Tracingrdquo Apple Accessed May 19 2020 httpswwwapplecomcovid19contacttracing
ASTHO 2012 ldquoPublic Health Collection Use Sharing and Protection of Informationrdquo Issue Brief ASTHO Legal Preparedness Series Association of State and Territorial Health Officials httpswwwasthoorgProgramsPreparednessPublic-Health-Emergency-LawPublic-Health-and- Information-Sharing-ToolkitCollection-Use-Sharing-and-Protection- Issue-Brief
Auxier Brooke Lee Rainie Monica Anderson Andrew Perrin Madhu Kumar and Erica Turner 2019 ldquoAmericans and Privacy Concerned Confused and Feeling Lack of Control Over Their Personal Informationrdquo Pew Research Center Internet Science amp Tech (blog) November 15 2019 httpswwwpewresearchorginternet20191115americans-and-privacy- concerned-confused-and-feeling-lack-of-control-over-their-personal-infor mation
Barth Susanne and Menno D T de Jong 2017 ldquoThe Privacy Paradox ndash Inves-tigating Discrepancies between Expressed Privacy Concerns and Actual Online Behavior ndash A Systematic Literature Reviewrdquo Telematics and Infor-matics 34 (7) 1038ndash58 httpsdoiorg101016jtele201704013
Bernstein Justin Taylor A Holroyd Jessica E Atwell Joseph Ali and Rupali J Limaye 2019 ldquoRockland Countyrsquos Proposed Ban against Unvaccinated Minors Balancing Disease Control Trust and Libertyrdquo Vaccine June httpsdoiorg101016jvaccine201905093
Berry Scott M Elizabeth A Petzold Peter Dull Nathan M Thielman Coleen K Cunningham G Ralph Corey Micah T McClain et al 2016 ldquoA Response Adaptive Randomization Platform Trial for Efficient Evalua-tion of Ebola Virus Treatments A Model for Pandemic Responserdquo Clinical Trials (London England) 13 (1) 22ndash30 httpsdoiorg101177 1740774515621721
Beukenhorst AL DM Schultz J McBeth R Lakshminarayana JC Sergeant and WG Dixon 2017 ldquoUsing Smartphones for Research Outside Clinical
Works Cited 129
Settings How Operating Systems App Evelopers and Users Determine Geolocation Data Quality in MHealth Studiesrdquo In MEDINFO 2017 Precision Healthcare through Infomatics 10ndash14 IOS Press httpebooks iospressnlvolumearticle48095
Bradshaw Tim 2020 ldquo2 Billion Phones Cannot Use Google and Apple Con-tact-Tracing Tech | Ars Technicardquo Ars Technica April 20 2020 httpsarstechnicacomtech-policy2020042-billion-phones-cannot-use-google-and-apple-contract-tracing-tech
Canca Cansu 2020 ldquoWhy lsquoMandatory Privacy-Preserving Digital Contact Tracingrsquo Is the Ethical Measure against COVID-19rdquo Medium May 4 2020 httpsmediumcomcansucancawhy-mandatory-privacy- preserving-digital-contact-tracing-is-the-ethical-measure-against- covid-19-a0d143b7c3b6
Cavalier Robert J ed 2011 Approaching Deliberative Democracy Theory and Practice Carnegie Mellon University Press
Cavoukian Ann 2010 ldquoPrivacy by Design The 7 Foundational Principlesrdquo Ontario Canada Information and Privacy Commissioner of Canada httpsiapporgmediapdfresource_centerpbd_implement_7found_ principlespdf
CDC 2005 ldquoVI Plan for Surveillance of Contacts of SARS Case Supplement B SARS Surveillancerdquo Centers for Disease Control and Prevention httpswwwcdcgovsarsguidanceb-surveillancecontactshtml
mdashmdashmdash 2020a ldquoContact Tracingrdquo Get and Keep America Open Supporting States Tribes Localities and Territories Centers for Disease Control and Prevention httpswwwcdcgovcoronavirus2019-ncovphpopen-americacontact-tracinghtml
mdashmdashmdash 2020b ldquoContact Tracing Part of a Multipronged Approach to Fight the COVID-19 Pandemicrdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention httpswwwcdcgovcoronavirus2019- ncovphpprinciples-contact-tracinghtml
mdashmdashmdash 2020c ldquoInterim Guidance for Businesses and Employersrdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention httpswwwcdcgovcoronavirus2019-ncovcommunityguidance-business- responsehtml
mdashmdashmdash 2020d ldquoOperational Considerations for the Identification of Healthcare Workers and Inpatients with Suspected COVID-19 in Non-US Healthcare Settingsrdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention httpswwwcdcgovcoronavirus2019-ncovhcpnon-us-settingsguidance-identify-hcw-patientshtml
130 Works Cited
mdashmdashmdash 2020e ldquoPreliminary Criteria for the Evaluation of Digital Contact Trac-ing Tools for COVID-19rdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention httpswwwcdcgovcoronavirus 2019-ncovdownloadsphpprelim-eval-criteria-digital-contact-tracingpdf
mdashmdashmdash 2020f ldquoLegal Authorities for Isolation and Quarantinerdquo Centers for Disease Control and Prevention httpswwwcdcgovquarantine aboutlawsregulationsquarantineisolationhtml
mdashmdashmdash 2020g ldquoDiscontinuation of Isolation for Persons with COVID-19 Not in Healthcare Settingsrdquo Interim Guidance Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention httpswww cdcgovcoronavirus2019-ncovhcpdisposition-in-home-patientshtml
mdashmdashmdash 2020h ldquoCOVID-19 Provisional CountsndashWeekly Updates by Select Demographic and Geographic Characteristicsrdquo CDC National Vital Statis-tics System httpswwwcdcgovnchsnvssvsrrcovid_weeklyindexhtm
Center for Survey Measurement (CSM) MEMORANDUM FOR Associate Directorate for Research and Methodology (ADRM) 2017 ldquoRespondent Confidentiality Concernsrdquo September 20 2017 httpswww2censusgovcacnacmeetings2017-11Memo-Regarding-Respondent-Confidentiality- Concernspdf
Cheng Hao-Yuan Shu-Wan Jian Ding-Ping Liu Ta-Chou Ng Wan-Ting Huang and Hsien-Ho Lin 2020 ldquoContact Tracing Assessment of COVID-19 Transmission Dynamics in Taiwan and Risk at Different Exposure Periods Before and After Symptom Onsetrdquo JAMA Internal Medicine May httpsdoiorg101001jamainternmed20202020
Cochrane Emily Claire Cain Miller and Jim Tankersley 2020 ldquoTrump Admin-istration Scales Back Paid Leave in Coronavirus Relief Lawrdquo The New York Times April 2 2020 sec US httpswwwnytimescom20200402uspoliticscoronavirus-paid-leavehtml
Crocker Andrew Kurt Opsahl and Bennett Cyphers 2020 ldquoThe Challenge of Proximity Apps For COVID-19 Contact Tracingrdquo Electronic Fron-tier Foundation April 10 2020 httpswwwefforgdeeplinks202004challenge-proximity-apps-covid-19-contact-tracing
Danquah Lisa O Nadia Hasham Matthew MacFarlane Fatu E Conteh Fatoma Momoh Andrew A Tedesco Amara Jambai David A Ross and Helen A Weiss 2019 ldquoUse of a Mobile Application for Ebola Contact Tracing and Monitoring in Northern Sierra Leone A Proof-of-Concept Studyrdquo BMC Infectious Diseases 19 (1) 810 httpsdoiorg101186s12879-019-4354-z
de Jong Bouke C Badou M Gaye Jeroen Luyten Bart van Buitenen Emman-
Works Cited 131
uel Andreacute Conor J Meehan Cian OrsquoSiochain et al 2019 ldquoEthical Considerations for Movement Mapping to Identify Disease Transmission Hotspotsrdquo Emerging Infectious Diseases 25 (7) httpsdoiorg103201eid2507181421
Devakumar Delan Geordan Shannon Sunil S Bhopal and Ibrahim Abu-bakar 2020 ldquoRacism and Discrimination in COVID-19 Responsesrdquo Lancet (London England) 395 (10231) 1194 httpsdoiorg101016S0140-6736(20)30792-3
Dixit Pranav 2020 ldquoIndiarsquos Contact Tracing App Is All But Mandatory So This Programmer Hacked It So That He Always Appears Saferdquo BuzzFeed News May 12 2020 httpswwwbuzzfeednewscomarticlepranavdixitindia-aarogya-setu-hacked
Doerr Megan Christine Suver and John Wilbanks 2016 ldquoDeveloping a Trans-parent Participant-Navigated Electronic Informed Consent for Mobile- Mediated Researchrdquo SSRN Scholarly Paper ID 2769129 Rochester NY Social Science Research Network httpsdoiorg102139ssrn2769129
Dredze Mark Michael J Paul Shane Bergsma and Hieu Tran 2013 ldquoCarmen A Twitter Geolocation System with Applications to Public Healthrdquo AAAI Workshops Workshops at the Twenty-Seventh AAAI Conference on Artifi-cial Intelligence httpswwwaaaiorgocsindexphpWSAAAIW13paperview70856497
Eckhoff Philip A and Andrew J Tatem 2015 ldquoDigital Methods in Epidemi-ology Can Transform Disease Controlrdquo International Health 7 (2) 77ndash78 httpsdoiorg101093inthealthihv013
EEOC 2020 ldquoWhat You Should Know About COVID-19 and the ADA the Rehabilitation Act and Other EEO Lawsrdquo US Equal Employment Opportunity Commission May 7 2020 httpswwweeocgovwyskwhat-you-should-know-about-covid-19-and-ada-rehabilitation-act-and-other-eeo-laws
Electronic Privacy Information Center Testimony to Congress 2020 ldquoEPIC to Congress Establish Privacy Safeguards for Digital Contact Tracingrdquo April 15 2020 httpsepicorgtestimonycongressEPIC-HEC-Contact-Tracing-Apr2020pdf
Elliott Douglas J Ana Kreacic Lorenzo Milans del Bosch and Lisa Quest 2020 ldquoData-Sharing in the Time of Coronavirusrdquo Oliver Wyman Forum Accessed May 19 2020 httpswwwoliverwymanforumcomfuture-of-data2020aprdata-sharing-in-the-time-of-coronavirushtml
Faden Ruth R and Tom L Beauchamp 1986 A History and Theory of Informed Consent Oxford University Press
132 Works Cited
FCC and FTC 2017 ldquoFCC-FTC Consumer Protection Memorandum of Understandingrdquo Federal Communications Commission and Federal Trade Commission httpswwwftcgovsystemfilesdocumentscooperation_agreements151116ftcfcc-moupdf
Ferretti Luca Chris Wymant Michelle Kendall Lele Zhao Anel Nurtay Lucie Abeler-Doumlrner Michael Parker David Bonsall and Christophe Fraser 2020 ldquoQuantifying SARS-CoV-2 Transmission Suggests Epidemic Control with Digital Contact Tracingrdquo Science 368 (6491) httpsdoiorg101126scienceabb6936
Fishkin James S and Peter Laslett eds 2003 Debating Deliberative Democ-racy Blackwell Publisher Ltd httpsonlinelibrarywileycomdoibook 1010029780470690734
Flanagan Mary Daniel C Howe and Helen Nissenbaum 2008 ldquoEmbodying Values in Technology Theory and Practicerdquo In Information Technology and Moral Philosophy edited by Jeroen van den Hoven and John Weckert 322ndash53 Cambridge University Press
Fraccaro Paolo Anna Beukenhorst Matthew Sperrin Simon Harper Jasper Palmier-Claus Shocircn Lewis Sabine N Van der Veer and Niels Peek 2019 ldquoDigital Biomarkers from Geolocation Data in Bipolar Disorder and Schizophrenia A Systematic Reviewrdquo Journal of the American Medical Informatics Association 26 (11) 1412ndash20 httpsdoiorg101093jamiaocz043
Fraser Christophe Lucie Abeler-Doumlrner Luca Ferretti Michael Parker Michelle Kendall and David Bonsall 2020 ldquoDigital Contact Tracing Comparing the Capabilities of Centralised and Decentralised Data Architectures to Effectively Suppress the COVID-19 Epidemic Whilst Maximising Freedom of Movement and Maintaining Privacyrdquo httpsgithubcomBDI-pathogenscovid-19_instant_tracingblobmasterCentralised20and20decen-tralised20systems20for20contact20tracingpdf
FTC 2012 ldquoProtecting Consumer Privacy in an Era of Rapid Changerdquo FTC Report Recommendations for Businesses and Policymakers Federal Trade Commission httpswwwftcgovsitesdefaultfilesdocumentsreports federal-trade-commission-report-protecting-consumer-privacy- era-rapid-change-recommendations120326privacyreportpdf
mdashmdashmdash 2020 ldquoPrivacy amp Data Security Update 2019rdquo Federal Trade Commis-sion httpswwwftcgovsystemfilesdocumentsreportsprivacy-data- security-update-20192019-privacy-data-security-report-508pdf
Furlanello Cesare Stefano Merler Stefano Menegon Sebastiano Mancuso and Gianni Bertiato 2002 ldquoNew WEBGIS Technologies for Geo-Location of
Works Cited 133
Epidemiological Data An Application for the Surveillance of the Risk of Lyme Borreliosis Diseaserdquo Giornale Italiano Di Aritmologia e Cardiosti-molazione Special Issue Proceedings of the 15th Int Congress ldquoThe New Frontiers of Arrhythmiasrdquo 5 (1) 241ndash45
Gan Nectar and David Culver 2020 ldquoChina Is Fighting the Coronavirus with a Digital QR Code Herersquos How It Worksrdquo CNN Business April 16 2020 httpswwwcnncom20200415asiachina-coronavirus-qr-code-intl-hnkindexhtml
Gibson Dustin G Adaeze C Wosu George William Pariyo Saifuddin Ahmed Joseph Ali Alain B Labrique Iqbal Ansary Khan Elizeus Rutebemberwa Meerjady Sabrina Flora and Adnan A Hyder 2019 ldquoEffect of Airtime Incentives on Response and Cooperation Rates in Non-Communicable Disease Interactive Voice Response Surveys Randomised Controlled Trials in Bangladesh and Ugandardquo BMJ Global Health 4 (5) e001604 httpsdoiorg101136bmjgh-2019-001604
Giglio Mike 2020 ldquoWould You Sacrifice Your Privacy to Get Out of Quarantinerdquo The Atlantic April 22 2020 httpswwwtheatlanticcompoliticsarchive202004coronavirus-pandemic-privacy-civil-liberties-911 609172
Guariglia Matthew 2020 ldquoThe Dangers of COVID-19 Surveillance Proposals to the Future of Protestrdquo Electronic Frontier Foundation April 29 2020 httpswwwefforgdeeplinks202004some-covid-19-surveillance-propos als-could-harm-free-speech-after-covid-19
Hadavas Chloe 2020 ldquoHow Effective Are Contact Tracing Appsrdquo Slate Maga-zine May 13 2020 httpsslatecomtechnology202005contact- tracing-apps-less-effective-icelandhtml
Hamilton Isobel Asher 2020 ldquoPoland Made an App That Forces Coronavirus Patients to Take Regular Selfies to Prove Theyrsquore Indoors or Face a Police Visitrdquo Business Insider March 23 2020 httpswwwbusinessinsidercompoland-app-coronavirus-patients-mandaotory-selfie-2020-3
Hargittai Eszter Minh Hao Nguyen Jaelle Fuchs Jonathan Gruber Will Marler Amanda Hunsaker and Goumlkccedile Karaoglu 2020 ldquoCovid-19 Study on Digital Media and the Coronavirus Pandemicrdquo Internet Use and Society Division Institute of Communication and Media Research University of Zurich httpwebuseorgcovid
Hargittai Eszter and Elissa Redmiles 2020 ldquoWill Americans Be Willing to Install COVID-19 Tracking Appsrdquo Scientific American Blog Net-work April 28 2020 httpsblogsscientificamericancomobservationswill-americans-be-willing-to-install-covid-19-tracking-apps
134 Works Cited
Hart Vi Divya Siddarth Bethan Cantrell Lila Tretikov Peter Eckersley John Langford Scott Leibrand et al 2020 ldquoOutpacing the Virus Dig-ital Response to Containing the Spread of COVID-19 While Mitigat-ing Privacy Risksrdquo Whitepaper 5 COVID-19 Rapid Response Impact Initiative Edmond J Safra Center for Ethics httpsdrivegooglecomfiled1vIN2AX-DDNW-S0aHq8xs0RJ2jkR_CckXview
Hemming K T P Haines P J Chilton A J Girling and R J Lilford 2015 ldquoThe Stepped Wedge Cluster Randomised Trial Rationale Design Anal-ysis and Reportingrdquo BMJ 350 (February) httpsdoiorg101136bmjh391
Hendrix Steve and Ruth Eglash 2020 ldquoIsrael Is Using Cellphone Surveil-lance to Warn Citizens You May Already Be Infectedrdquo Washington Post Accessed May 19 2020 httpswwwwashingtonpostcomworldmiddle_eastisrael-is-using-cellphone-surveillance-to-warn-citizens-you-may- already-be-infected2020031968267294-69e7-11ea-b199-3a9799c54 512_storyhtml
Heneghan Carl Jon Brassey and Tom Jefferson 2020 ldquoCOVID-19 What Proportion Are Asymptomaticrdquo Centre for Evidence-Based Medi-cine httpswwwcebmnetcovid-19covid-19-what-proportion-are- asymptomatic
HHS 2019 ldquoPublic Health Emergency Declarationrdquo Public Health Emergency Accessed May 19 2020 httpswwwphegovPreparednesslegalPagesphedeclarationaspx
Hickey Matt 2014 ldquoCarriers Can Now Install Apps On Android Handsets Without Customersrsquo Permissionrdquo Forbes December 1 2014 httpswww forbescomsitesmatthickey20141201carriers-can-now-install- apps-on-android-handsets-without-customers-permission
Hinch Robert Will Probert Anel Nurtay Michelle Kendall Chris Wymant Matthew Hall Katrina Lythgoe et al 2020 ldquoEffective Configurations of a Digital Contact Tracing App A Report to NHSXrdquo fileUsersameliahoodDownloadsReport20-20Effective20App20Configurations20(1)pdf
Ingram David 2020 ldquoApple Google Push Makers of Coronavirus Apps Not to Record User Locationrdquo May 4 2020 NBC News Accessed May 19 2020 httpswwwnbcnewscomtechtech-newscoronavirus-apps-won-t-be-able- record-users-location-apple-n1199586
Iwaya Leonardo H Jane Li Simone Fischer-Huumlbner Rose-Mharie Aringhlfeldt and Leonardo A Martucci 2019 ldquoE-Consent for Data Privacy Consent Management for Mobile Health Technologies in Public Health Surveys and
Works Cited 135
Disease Surveillancerdquo Studies in Health Technology and Informatics 264 (August) 1223ndash27 httpsdoiorg103233SHTI190421
ldquoJoint Statement on Contact Tracingrdquo 2020 April 19 httpscryptobriefingcomwp-contentuploads202004Joint-Statement-from-Researcherspdf
Kahn Gilmor Daniel 2020 ldquoPrinciples for Technology-Assisted Contact- Tracingrdquo White Paper American Civil Liberties Union httpswwwaclu orgreportaclu-white-paper-principles-technology-assisted-contact-tracing
Kim Max S 2020 ldquoSeoulrsquos Radical Experiment in Digital Contact Tracingrdquo The New Yorker April 17 2020 httpswwwnewyorkercomnewsnews-deskseouls-radical-experiment-in-digital-contact-tracing
Kim Nemo 2020 ldquolsquoMore Scary than Coronavirusrsquo South Korearsquos Health Alerts Expose Private Livesrdquo The Guardian March 6 2020 sec World news httpswwwtheguardiancomworld2020mar06more-scary- than-coronavirus-south-koreas-health-alerts-expose-private-lives
Kirzinger Ashley Liz Hamel Cailey Muntildeana Audrey Kearney and Mollyann Brodie 2020 ldquoKFF Health Tracking Poll ndash Late April 2020 Coronavirus Social Distancing and Contact Tracingrdquo Kaiser Family Foundation (blog) April 24 2020 httpswwwkfforgcoronavirus-covid-19issue-briefkff-health-tracking-poll-late-april-2020
Knobel Cory and Geoffrey C Bowker 2011 ldquoComputing Ethics Value in Designrdquo Communications of the ACM 54 (7) 26ndash28 httpsdoiorgdoi10114519657241965735
Lee Ramon Rosa R Cui Kathryn E Muessig Harsha Thirumurthy and Joseph D Tucker 2014 ldquoIncentivizing HIVSTI Testing A Systematic Review of the Literaturerdquo AIDS and Behavior 18 (5) 905ndash12 https doiorg101007s10461-013-0588-8
Leprince-Ringuet Daphne 2020 ldquoContact-Tracing Apps Why the NHS Said No to Apple and Googlersquos Planrdquo ZDNet April 28 2020 httpswwwzdnetcomarticlecontact-tracing-apps-why-the-nhs-said-no-to-apple-and- googles-plan
Lovejoy Ben 2020 ldquoNHS Contact Tracing App Hands-on as the Test Goes Liverdquo 9to5Mac (blog) May 7 2020 https9to5maccom20200507nhs-contact-tracing
Mahmood Sultan Khaled Hasan Michelle Colder Carras and Alain Labri-que 2020 ldquoGlobal Preparedness Against COVID-19 We Must Leverage the Power of Digital Healthrdquo JMIR Public Health and Surveillance 6 (2) e18980 httpsdoiorg10219618980
Mathews Simon C Michael J McShea Casey L Hanley Alan Ravitz Alain B Labrique and Adam B Cohen 2019 ldquoDigital Health A Path to
136 Works Cited
Validationrdquo Npj Digital Medicine 2 (1) 1ndash9 httpsdoiorg101038s41746-019-0111-3
Mello By Michelle M and C Jason Wang 2020 ldquoEthics and Governance for Digital Disease Surveillancerdquo Science May httpsdoiorg101126science abb9045
Mills Rodrigo Chris 2020 ldquoDigital Contact Tracing Is Becoming Available but Is It Effectiverdquo The Hill May 7 2020 httpsthehillcompolicytechnolo-gy496498-digital-contact-tracing-is-becoming-available-but-is-it-effective
MIT nd ldquoPrivate Kit Safe Paths Privacy-by-Design Covid19 Solutions Using GPS+Bluetooth for Citizens and Public Health Officialsrdquo Safepaths Accessed May 20 2020 httpssafepathsmitedu
ldquoMobile Cellular Subscriptions (per 100 People)rdquo 2018 World Telecommuni-cationICT Development Report International Telecommunication Union httpsdataworldbankorgindicatorITCELSETSP2
Moore Sarah Anne-Marie Tasseacute Adrian Thorogood Ingrid Winship Marsquon Zawati and Megan Doerr 2017 ldquoConsent Processes for Mobile App Mediated Research Systematic Reviewrdquo JMIR MHealth and UHealth 5 (8) e126 httpsdoiorg102196mhealth7014
Morse Jack 2020 ldquoNorth Dakota Launched a Contact-Tracing App Itrsquos Not Going Wellrdquo Mashable Accessed May 19 2020 httpsmashablecomarticlenorth-dakota-contact-tracing-app
Muller Robert T 2020 ldquoCOVID-19 Brings a Pandemic of Conspiracy Theo-riesrdquo Psychology Today April 24 2020 httpswwwpsychologytodaycomblogtalking-about-trauma202004covid-19-brings-pandemic-conspiracy- theories
Mulligan Stephen P and Chris D Linebaugh 2019 ldquoData Protection Law An Overviewrdquo R45631 Congressional Research Service httpscrsreports congressgovproductpdfRR45631
National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research 1979 ldquoThe Belmont Report Ethical Principles and Guidelines for the Protection of Human Subjects of Researchrdquo US Dept of Health amp Human Services Office of Human Research Protections httpswwwhhsgovohrpregulations-and-policybelmont-reportread-the-belmont-reportindexhtml
NDDoH 2020 ldquoNorth Dakota Launches Care19 App to Combat COVID-19rdquo North Dakota Department of Health April 7 2020 httpswwwhealth ndgovnewsnorth-dakota-launches-care19-app-combat-covid-19
NYC DOH 2020 ldquoAge-Adjusted Rates of Lab Confirmed COVID-19 Non-hospitalized Cases Estimated Non-Fatal Hospitalized Cases and Persons Known to Have Died per 100000 by RaceEthnicity Grouprdquo New York
Works Cited 137
City Department of Health httpswww1nycgovassetsdohdownloadspdfimmcovid-19-deaths-race-ethnicity-04242020-1pdf
OrsquoRielly Michael 2018 ldquoFCC Regulatory Free Arenardquo Federal Communica-tions Commission June 1 2018 httpswwwfccgovnews-eventsblog 20180601fcc-regulatory-free-arena
OSHA 2020 ldquoGuidance on Preparing Workplaces for COVID-19rdquo OSHA 3990-03 2020 Department of Labor Occupational Safety and Health Act of 1970 httpswwwoshagovPublicationsOSHA3990pdf
Pallmann Philip Alun W Bedding Babak Choodari-Oskooei Munyaradzi Dimairo Laura Flight Lisa V Hampson Jane Holmes et al 2018 ldquoAdaptive Designs in Clinical Trials Why Use Them and How to Run and Report Themrdquo BMC Medicine 16 (1) 29 httpsdoiorg101186s12916-018-1017-7
Pew Research Center 2017 ldquoUS Muslims Concerned About Their Place in Society but Continue to Believe in the American Dreamrdquo Religion amp Public Life Pew Research Center httpswwwpewforumorg20170726findings-from-pew-research-centers-2017-survey-of-us-muslims
mdashmdashmdash 2020 ldquoDemographics of Mobile Device Ownership and Adoption in the United Statesrdquo Internet amp Technology Pew Research Center Accessed May 19 2020 httpswwwpewresearchorginternetfact-sheetmobile
PHLS 2002 ldquoPrinciples of the Ethical Practice of Public Healthrdquo Public Health Leadership Society httpswwwaphaorg-mediafilespdfmembergroupsethicsethics_brochureashx
PIH 2020a ldquoPart I Testing Contact Tracing and Community Management of COVID-19rdquo PIH Guide | COVID-19 STOP COVID Partners in Health httpswwwpihorgsitesdefaultfiles2020-04PIH_Guide_COVID_Part_I_Testing_Tracing_Community_Managment_4_4pdf
mdashmdashmdash 2020b ldquoBox It Inrdquo PIH Guide | COVID-19 Partners in Health httpspreventepidemicsorgcovid19resourcesbox-it-in
Resolve to Save Lives 2020 ldquoCOVID-19 Contact Tracing Playbookrdquo Vital Strategies httpscontacttracingplaybookresolvetosavelivesorg
Reston Maeve Kristina Sgueglia and Cheri Mossburg 2020 ldquoGovernors on East and West Coasts Form Pacts to Decide When to Reopen Economiesrdquo CNN Politics April 13 2020 httpswwwcnncom20200413politicsstates-band-together-reopening-plansindexhtml
Rithalia Amber Catriona McDaid Sara Suekarran Lindsey Myers and Amanda Sowden 2009 ldquoImpact of Presumed Consent for Organ Donation on Donation Rates A Systematic Reviewrdquo BMJ 338 (January) httpsdoiorg101136bmja3162
Rodrigues Rafaela Alina Husain Amanda Couture-Carron Leslye E Orloff
138 Works Cited
and Nawal H Ammar 2018 ldquoPromoting Access to Justice for Immigrant and Limited English Proficient Crime Victims in an Age of Increased Immi-gration Enforcement Initial Report from a 2017 National Surveyrdquo Wash-ington DC National Immigrant Womenrsquos Advocacy Project American University Washington College of Law httpniwaplibrarywclamericaneduwp-contentuploadsImmigrant-Access-to-Justice-National-Reportpdf
Simpson Erin and Adam Conner 2020 ldquoDigital Contact Tracing to Contain the Coronavirusrdquo Center for American Progress httpswww americanprogressorgissuestechnology-policynews20200422483521digital-contact-tracing-contain-coronavirus
Singer Eleanor and Cong Ye 2013 ldquoThe Use and Effects of Incentives in Surveysrdquo The ANNALS of the American Academy of Political and Social Science 645 (1) 112ndash41 httpsdoiorg1011770002716212458082
Thornton Rebecca L 2008 ldquoThe Demand for and Impact of Learning HIV Statusrdquo The American Economic Review 98 (5) 1829ndash63 httpsdoi org101257aer9851829
US DOE 2009 ldquoState Regulation of Private Schoolsrdquo US Dept of Education Office of Non-Public Education httpswww2edgovadminscommchoiceregprivschlregprivschlpdf
Valentino-DeVries Jennifer 2020 ldquoCellphone Carriers Face $200 Million Fine for Not Protecting Location Datardquo The New York Times February 28 2020 sec Technology httpswwwnytimescom20200228technologyfcc-cellphones-location-data-fineshtml
Valentino-DeVries Jennifer Natasha Singer and Aaron Krolik 2020 ldquoA Scramble for Virus Apps That Do No Harmrdquo The New York Times April 29 2020 sec Business httpswwwnytimescom20200429businesscoronavirus-cellphone-apps-contact-tracinghtml
Washington PostndashUMD ldquoWashington Post-University of Maryland National Pollrdquo 2020 Washington Post April 21 2020 httpswwwwashingtonpostcomcontextwashington-post-university-of-maryland-national-poll-april-21-26-20203583b4e9-66be-4ed6-a457-f6630a550ddf
Watson Crystal Anita Cicero James Blumenstock and Michael Fraser 2020 ldquoA National Plan to Enable Comprehensive COVID-19 Case Finding and Contact Tracing in the USrdquo Johns Hopkins Bloomberg School of Public Health Center for Health Security and the Association of State and Terri-torial Health Officials httpswwwcenterforhealthsecurityorgour-workpubs_archivepubs-pdfs2020200410-national-plan-to-contact-tracingpdf
Works Cited 139
WHO 2017 ldquoWHO Guidelines on Ethical Issues in Public Health Surveillancerdquo World Health Organization httpswwwwhointethicspublicationspublic-health-surveillanceen
mdashmdashmdash 2020 ldquoCoronavirus Disease 2019 (COVID-19)rdquo Situation Report 73 World Health Organization httpswwwwhointdocsdefault-sourcecoro-navirusesituation-reports20200402-sitrep-73-covid-19pdf
- _iy90yrtqgtxz
- _GoBack
- Acknowledgments
- Preface
- Lead Authors and Contributors
- Acronyms and Abbreviations
- Summary
-
- Introduction
- DCTT Features Functions and Potential Applications
- Summary of Recommendations
- Summary of Analysis
-
- Introduction
-
- Guiding Principles for the Use of Digital Public Health Technologies for Pandemic Response
-
- Public Health Perspective
-
- Types of Information Collected through Contact Tracing
- How Contact Tracing Information Informs Public Health Action
- Characteristics That Make Data Useful to Public Health for Reducing Disease Transmission
-
- Digital Technology and Contact Tracing
-
- Characteristics of SARS-CoV-2 Relevant to Candidate Digital Solutions
- Previously Existing Contact Tracing Technologies
- Introduction of Novel Digital Contact Tracing Technologies
- Relevant Differences between Manual and Digital Contact Tracing
-
- Ethics of Designing and Using DCTT
-
- Justifying the Use of DCTT Systems
- Monitoring and Evaluating Technologies to Inform Policy and Practice
- Public Trust and Public Attitudes
- Designing Flexible Technology to Maximize Public Health Utility While Respecting Other Values
- Policy Positions to Advance Widespread Use of Digital Contact Tracing Technologies
- Disclosure and AuthorizationConsent
- Promoting Equity and Fairness
- Instituting Transparent Governance and Oversight
-
- Legal Considerations
-
- Data Privacy and Data Security Laws
- Health Information Privacy
- Labor and Employment Privacy Rights
- Constitutional Privacy Rights
- Consent
- Anti-discrimination and Individual Freedom Laws
-
- Recommendations
-
- Public Health
- Ethics
- Legislative
-
- Resources
-
- US Government Response
- Other Governmental and Nongovernmental Organizations
- Digital Contact Tracing Experiences from Other Countries
- Specific Digital ProductsApps
- Polling
- Popular Press
- Commentaries
- Academic Literature
-
- Works Cited
-
DIGITAL CONTACT TRACING FOR PANDEMIC RESPONSEEthics and Governance Guidance
Johns Hopkins University PressBaltimore
Edited by Jeffrey P Kahn PhD MPH Chair
Johns Hopkins Project on Ethics and Governance of Digital Contact Tracing Technologies
copy 2020 Johns Hopkins University Press
This work is also available in an Open Access edition which is licensed
under a Creative Commons AttributionndashNonCommercialndashNoDerivatives 40
International License httpscreativecommonsorglicensesby-nc-nd40
All rights reserved Published 2020
Printed in the United States of America on acid-free paper
9 8 7 6 5 4 3 2 1
Johns Hopkins University Press
2715 North Charles Street
Baltimore Maryland 21218-4363
wwwpressjhuedu
Library of Congress Cataloging-in-Publication Data is available
ISBN 978-1-4214-4061-3 (paperback acid-free paper)
ISBN 978-1-4214-4062-0 (ebook)
ISBN 978-1-4214-4063-7 (ebook open access)
Special discounts are available for bulk purchases of this book For more information
please contact Special Sales at specialsalespressjhuedu
Johns Hopkins University Press uses environmentally friendly book materials
including recycled text paper that is composed of at least 30 percent post-
consumer waste whenever possible
Contents
Lead Authors and Contributors vii
Preface ix
Acknowledgments xi
Acronyms and Abbreviations xiii
Summary 1
Introduction 1
DCTT Features Functions and Potential Applications 3
Summary of Recommendations 9
Summary of Analysis 13
Introduction 23
Guiding Principles for the Use of Digital Public Health
Technologies for Pandemic Response 25
1 Public Health Perspective 29
Types of Information Collected through Contact Tracing 29
How Contact Tracing Information Informs Public Health Action 31
Characteristics That Make Data Useful to
Public Health for Reducing Disease Transmission 32
2 Digital Technology and Contact Tracing 35
Characteristics of SARS-CoV-2 Relevant to Candidate Digital Solutions 35
Previously Existing Contact Tracing Technologies 36
Introduction of Novel Digital Contact Tracing Technologies 37
Relevant Differences between Manual and Digital Contact Tracing 41
vi Contents
3 Ethics of Designing and Using DCTT 43
Justifying the Use of DCTT Systems 44
Monitoring and Evaluating Technologies to Inform Policy and Practice 45
Public Trust and Public Attitudes 48
Designing Flexible Technology to Maximize Public Health Utility
While Respecting Other Values 50
Policy Positions to Advance Widespread Use of Digital Contact
Tracing Technologies 59
Disclosure and AuthorizationConsent 63
Promoting Equity and Fairness 69
Instituting Transparent Governance and Oversight 72
4 Legal Considerations 75
Data Privacy and Data Security Laws 77
Health Information Privacy 82
Labor and Employment Privacy Rights 84
Constitutional Privacy Rights 86
Consent 93
Anti-discrimination and Individual Freedom Laws 94
5 Recommendations 97
Public Health 97
Ethics 98
Legislative 102
Resources 103
US Government Response 103
Other Governmental and Nongovernmental Organizations 106
Digital Contact Tracing Experiences from Other Countries 112
Specific Digital ProductsApps 115
Polling 117
Popular Press 119
Commentaries 120
Academic Literature 121
Works Cited 127
Lead Authors and Contributors
vii
Lead Authors
Joseph Ali JD Assistant Professor Department of International Health Johns Hopkins Bloomberg School of Public Health Core Faculty amp Associate Director for Global Programs Johns Hopkins Berman Institute of Bioethics
Anne Barnhill PhD Core Faculty amp Research Scholar Johns Hopkins Berman Institute of Bioethics
Anita Cicero JD Deputy Director Johns Hopkins Center for Health Security Visiting Faculty Johns Hopkins Bloomberg School of Public Health
Katelyn Esmonde PhD Hecht-Levi Postdoctoral Fellow Johns Hopkins Berman Institute of Bioethics
Amelia Hood MA Research Program Coordinator Johns Hopkins Berman Insti-tute of Bioethics
Brian Hutler PhD JD Hecht-Levi Postdoctoral Fellow Johns Hopkins Berman Institute of Bioethics
Jeffrey Kahn PhD MPH Andreas C Dracopoulos Director Johns Hopkins Ber-man Institute of Bioethics
Alan Regenberg MBE Director of Outreach amp Research Support Associate Fac-ulty Johns Hopkins Berman Institute of Bioethics
Crystal Watson DrPH MPH Senior Scholar Johns Hopkins Center for Health Security Assistant Professor Department of Environmental Health and Engi-neering Johns Hopkins Bloomberg School of Public Health
Matthew Watson Senior Analyst Johns Hopkins Center for Health Security Senior Research Associate Department of Environmental Health and Engi-neering Johns Hopkins Bloomberg School of Public Health
Other Contributors
Robert Califf MD MACC Head of Clinical Policy and Strategy Verily and Goo-gle Health
Ruth Faden PhD MPH Philip Franklin Wagley Professor of Biomedical Ethics amp Founder Johns Hopkins Berman Institute of Bioethics
viii Lead Authors and Contributors
Divya Hosangadi MSPH Senior Analyst Johns Hopkins Center for Health Secu-rity Research Associate Department of Environmental Health and Engi-neering Johns Hopkins Bloomberg School of Public Health
Nancy Kass ScD Deputy Director for Public Health amp Phoebe R Berman Pro-fessor of Bioethics and Public Health Johns Hopkins Berman Institute of Bioethics
Alain Labrique PhD MHS MS Professor Department of International Health Johns Hopkins Bloomberg School of Public Health Director Johns Hopkins Global Health Initiative
Deven McGraw JD MPH LLM Chief Regulatory Officer CiitizenMichelle Mello JD PhD Professor of Law Stanford Law School Professor of
Health Research and Policy Stanford University School of MedicineMichael Parker BEd (Hons) MA PhD Director Wellcome Centre for Ethics and
Humanities Ethox Centre University of OxfordStephen Ruckman JD MSc MAR Senior Advisor to the President for Policy
Office of the President Johns Hopkins UniversityLainie Rutkow JD MPH PhD Senior Advisor to the President for National Cap-
ital Academic Strategy Office of the President Johns Hopkins UniversityJosh Sharfstein MD Vice Dean for Public Health Practice and Community
Engagement Professor of the Practice Johns Hopkins Bloomberg School of Public Health
Jeremy Sugarman MD MPH MA Deputy Director for Medicine Harvey M Meyerhoff Professor of Bioethics and Medicine Johns Hopkins Berman Institute of Bioethics Department of Medicine Johns Hopkins School of Medicine and Department of Health Policy and Management Johns Hop-kins Bloomberg School of Public Health
Eric Toner MD Senior Scholar Johns Hopkins Center for Health Security Senior Scientist Department of Environmental Health and Engineering Johns Hop-kins Bloomberg School of Public Health
Marc Trotochaud MSPH Analyst Johns Hopkins Center for Health Security Research Associate Department of Environmental Health and Engineering Johns Hopkins Bloomberg School of Public Health
Effy Vayena PhD Professor Health Ethics amp Policy Lab Department of Health Sciences amp Technology ETH Zurich
Tal Zarsky JSD LLM LLB Professor of Law University of Haifa Faculty of Law Visiting Scholar University of Pennsylvania Law School (2019ndash2020)
Preface
ix
Digital technologies are being developed and promoted to support the public health response to the COVID-19 pandemic with discussion and implementation planning in the United States by localities states institu-tions and employers Key decision makers and stakeholdersmdashincluding government officials institutional leaders employers digital technology developers and the publicmdashrequire clear and well-supported guidance to inform the deployment and use of these technologies as well as of the data they collect store and share While technology-based approaches are currently unable to provide solutions on their own experiences in other countries indicate that they could be used successfully in conjunc-tion with traditional and novel public health methods
This report reflects a rapid research and expert consensus group ef-fort led by the Berman Institute of Bioethics and the Center for Health Se-curity at Johns Hopkins University It draws on experts from both inside and outside Johns Hopkins in bioethics health security public health technology development engineering public policy and law The report highlights issues that must be addressed and provides recommendations for the use of digital technologies as part of contact tracing
The analysis offered here is focused on answering the following questions
bull Can digital contact tracing technologies (DCTT) be effective as part of public health responses to the pandemic and if so to what degree for which specific types of functions with what confidence and with what requirements
bull How can these technologies serve the interests of public health while respecting other individual and collective interests such as ensuring equitable distribution of benefits and burdens and limit-ing infringement on privacy and other civil liberties
x Preface
bull What are the ethical legal policy and governance guardrails cur-rently in place around such technologies and what else is needed
bull What additional guardrails are required to ensure that the goals of public health in using these technologies are achievable in ways that are ethically and legally sound
To answer these questions the report examines some core aspects of dig-ital technologies applied to contact tracing focusing on
bull the value of and basic methods for traditional public health surveil-lance and contact tracing
bull candidate technological products to enhance public health surveil-lance and contact tracing how they work and their comparative value for public health
bull core ethical legal and governance considerations and how they relate to relevant features of candidate technological solutions and
bull what is needed to move forward responsibly with the use of digital technology in support of public health surveillance acknowledg-ing gaps in our current understanding
The project involved in-depth analysis by a dedicated team of faculty postdoctoral fellows and research staff working over the course of only a few weeks but with great intensity drafting a report in collaboration with 26 total contributors writing commenting and revising through multiple drafts with the penultimate draft ldquopressure-testedrdquo by review and discus-sion at a virtual workshop of invited experts and stakeholders held on May 13 2020 and the final version completed on May 21 2020 The report and analysis builds on the excellent work of others in some parts of this territory while focusing on the gaps in analysis and areas that have not been sufficiently addressed The goal is to offer comprehensive guid-ance to relevant stakeholders to advance public health response during the COVID-19 pandemic Given the rapidly evolving territory into which DCTT is being introduced this report will by necessity be something of a living document updated as often as information dictates in order to continue to offer leading-edge analysis and guidance Versions will be noted in the digital and print editions
Acknowledgments
xi
Efforts like this project require teams and even small armies to be carried out successfully and this was no exception except that it was many fewer people working many more hours than could reasonably be expected of them From the initial kernel of an idea to the publication of this report in book form this project took just over a month total That seems impossi-ble even as I know it is accurate and it speaks to the incredible commit-ment hard work research skills and analytic acumen of our colleagues at Johns Hopkinsmdashthe core team are deservedly listed as lead authors of this report
None of this would have been possible without the supportmdashmoral and financialmdashand encouragement of Johns Hopkins University Presi-dent Ronald J Daniels who was the first to suggest the idea to me of taking on this topic He provided not only support and encouragement but the imprimatur of his office including help guidance and counsel from Prof Lainie Rutkow senior advisor to the president Lainie played a more integral role than that description captures reflected in part by her inclusion among the reportrsquos contributors but she deserves special acknowledgement for shepherding us through to the end
I mentioned that this was a team effort and every team requires an effective leader My colleague Prof Joseph Ali stepped into that role as we undertook the project and then he worked seven days a week along with the rest of the core research and writing team always unfailingly positive and deeply engaged in the work He along with Prof Anne Barnhill Alan Regenberg Amelia Hood and Drs Katelyn Esmonde Brian Hutler and Crystal Watson all deserve special thanks for doing so much in so little time all while working under the grinding social distancing restrictions of the 2020 pandemic That work was supported by Arnold amp Porter Kaye Scholer LLP with legal research and other assistancemdasha huge thanks to
xii Acknowledgments
them Finally the 16 contributing authors were incredibly generous with their time energy and insights all on ridiculously tight timelines and never a complaint or objection
The project benefited greatly from a number of experts who provided written feedback on drafts and who attended the virtual workshop to test our recommendations including Miles Stewart Rob Nichols Smisha Aagarwal Karl Steiner Anupam Joshi Charles Scheeler Ford Rowan and Jay Wagley
Last the fact that this report appears in published book form by Johns Hopkins University Press is another minor miracle from manu-script to printed book in under a week Thanks to JHUP Director Barbara Kline Pope and her team for being willing to take on the challenge and for the incredible focused effort it required
my heartfelt thanks andappreciation to you all
Jeffrey Kahn
Acronyms and Abbreviations
xiii
ADA Americans with Disabilities ActBLE Bluetooth Low EnergyCalOPPA California Online Privacy Protection ActCBP Customs and Border ProtectionCCPA California Consumer Privacy ActCDC Centers for Disease Control and PreventionCLOUD Act Clarifying Lawful Overseas Use of Data ActCOPPA Childrenrsquos Online Privacy Protection ActCOV+ confirmed positive SARS-CoV-2 test resultCOVID-19 coronavirus disease 2019CPNI customer proprietary network informationCSLI cell-site location informationDCTT digital contact tracing technology and closely related
digital health productsECPA Electronic Communications Privacy ActEEOC Equal Employment Opportunity CommissionE-SIGN Electronic Signatures in Global and National
Commerce ActEU European UnionFCC Federal Communications CommissionFTC Federal Trade CommissionFTCA Federal Trade Commission ActGIS geographic information systemGPS global positioning systemHHS US Department of Health and Human ServicesHIPAA Health Insurance Portability and Accountability ActHIV human immunodeficiency virusICU intensive care unit
xiv Acronyms and Abbreviations
IRB institutional review boardJHU Johns Hopkins UniversityOCR Office for Civil Rights US Department of Health amp
Human ServicesOSHA Occupational Safety and Health AdministrationPHI protected health informationPII personally identifiable informationPPE personal protective equipmentPPPT privacy-preserving proximity trackingQR Code quick response codeRFRA Religious Freedom Restoration ActRLUIPA Religious Land Use and Institutionalized Persons ActSARS-CoV-2 severe acute respiratory syndrome coronavirus 2SCA Stored Communications ActSTI sexually transmitted infection
DIGITAL CONTACT TRACING FOR PANDEMIC RESPONSE
11
Summary
Introduction
Public health professionals around the world are working tirelessly to respond to the COVID-19 pandemic using tried-and-tested public health methods for infectious disease surveillance and control These traditional methods are essential to the global COVID-19 response To complement these actions and potentially augment the speed and efficacy of the pub-lic health workforce digital technologies are being harnessed Given the scale of the pandemic significant efforts are being undertaken to develop and leverage public-facing and health-system-supportive technology solu-tions including smartphone apps and other digital tools that may aid public health surveillance and contact tracing
Digital contact tracing technology and closely related digital health products (together DCTT) have been used in several countries as part of broader disease surveillance and containment strategies In the United States DCTT has been proposed as an integral part of some plans to ldquoreopenrdquo the country (Allen et al 2020 Hart et al 2020 Simpson and Conner 2020) It is almost certain that these and related technologies will become part of not only the COVID-19 response but also the larger toolbox for future public health communicable disease prevention and control
These technologies have significant promise They also raise import-ant ethical legal and governance challenges that require comprehensive analysis in order to support decision-making Government officials pub-lic health leaders leaders of institutions employers digital technology de-velopers and the public all must be adequately informed in order to make
2 Digital Contact Tracing for Pandemic Response
responsible choices Johns Hopkins University recognized the importance of helping to guide this process It organized an expert group with mem-bers from inside and outside of Hopkins and led by its Berman Institute of Bioethics in collaboration with the Center for Health Security Its charge was to examine the ethics law policy and public health implications of using digital technologies as part of pandemic response and to develop guidance including a framework and actionable recommendations for governmental and institutional decision makers
Overall this expert group urges a stepwise approach that prioritizes align-ment of technology with public health needs and public values building choice into design architecture and capturing real-world results and impacts to allow adjustments as required Further we urge an approach that recog-nizes that there are complicated issues to resolve for governments insti-tutions and businesses and that introduction of DCTT must include public engagement and ongoing assessments to improve both performance and adoption
Specific recommendations include the following
bull There is no ldquoone size fits allrdquo approach to DCTT Technology design should not be static but it should be capable of evolving depending upon local conditions new evidence and changing preferences and priorities
bull Technology companies alone should not control the terms condi-tions or capabilities of DCTT nor should they presume to know what is acceptable to members of the public
bull DCTT should be designed to have a base set of features that pro-tect privacy with layers of additional capabilities that users may choose to activate An initial default should be that user location data are not shared but users should be provided with easy mech-anisms and prompts to allow for opting-in to this capability with encouragement to the public if it is shown to be critical to achieving public health goals
bull Data collected through DCTT should be made available to public health professionals and to researchers in de-identified form to support population-level epidemiologic analyses
Summary 3
bull Those who authorize use of DCTT within a particular jurisdic-tion or institution should continuously and systematically monitor the technologyrsquos performance in that context This should include monitoring for effectiveness and benefit monitoring for harms and monitoring for the fair distribution of both benefits and harms
bull Governments should not require mandatory use of DCTT given uncertainty about potential burdens and benefits Additional tech-nology user and real-world testing is needed
Through in-depth analysis and recommendations this report seeks to guide decision-making and enhance understanding of
bull the value of and basic methods for traditional public health surveil-lance and contact tracing
bull candidate technological products to enhance public health surveil-lance and contact tracing and their comparative value for public health
bull core ethical legal policy and governance considerations and how they relate to relevant features of candidate technological solutions and
bull what is needed to move forward responsibly with the use of digital technology in support of public health surveillance acknowledg-ing gaps in our current understanding
The full set of recommendations are intended to (1) support effective and informed adoption of DCTT (2) encourage design of flexible technol-ogies that maximize public health utility while respecting other values (3) establish meaningful processes for user disclosure and authorization (consent) (4) promote equity and fairness in the uses of DCTT and (5) foster transparent governance and oversight
DCTT Features Functions and Potential Applications
Digital contact tracing technologies and platforms can be roughly catego-rized into three broad approaches along a spectrum of potential policies and methods a maximal approach (typified by the South Korean govern-
4 Digital Contact Tracing for Pandemic Response
mentrsquos centralized and triangulated data collection (M S Kim 2020)) a minimal approach (typified by the AppleGoogle decentralized priva-cy-preserving proximity tracking (PPPT) and contact notification (Apple and Google nd)) and a diverse range of middle-ground approaches that aim to augment manual contact tracing with the collection of digital data that can be shared with public health authorities
Minimal approaches such as the AppleGoogle PPPT use Bluetooth Low Energy (BLE) ldquohandshakesrdquo that record close contact between mo-bile phone users but do not register the location in which the contact hap-pened In most architectures these proximity data are stored in the usersrsquo phones as anonymized ldquobeaconsrdquo that cannot be used to re-identify the users directly If a user with a PPPT app installed on their phone tests pos-itive and enters test results into their app those who have been in contact with them can be notified by the app This ldquoexposure notificationrdquo can be automatic or at the discretion of the COV+ person depending on the app design If notified a user who has been in contact with a COV+ individ-ual would receive a push notification alerting them to possible exposure (which may be timestamped) but with no other identifying information
The most prevalent middle-ground approach in the US context in-volves the collection and storage of personal datamdashincluding identifying information and location datamdashon the userrsquos phone These decentralized but personally identifiable data can then be voluntarily shared with pub-lic health officials if the user tests positive for SARS-CoV-2 (severe acute respiratory syndrome coronavirus 2) For example a team at the Mas-sachusetts Institute of Technology (MIT) has developed an app called Private Kit Safe Paths (MIT nd) that stores usersrsquo location data on their phone for 28 days If a user tests positive she can voluntarily upload her location data to a website that is accessible only to public health officials Officials can then analyze these personally identifiable data and subse-quently broadcast redacted and de-identified data to other users Healthy users would have access to these redacted location data of COV+ users but their own data would not leave their phones At a minimum the storage of user location data can function as a ldquomemory aidrdquo if the user tests positive but releasing the data to public health authorities may help to analyze the spread of SARS-CoV-2 and alert individuals or groups that have been in contact with COV+ patients
Summary 5
The US Centers for Disease Control and Prevention (CDC) has pub-lished preliminary criteria for evaluating capabilities and attributes of DCTT (CDC 2020e) These and other resources suggest that a compre-hensive assessment of DCTT and its potential to advance the publicrsquos health will require careful consideration of numerous interconnected fac-tors that interact in complex ways and must be navigated within the chal-lenging contexts of uncertainty and urgent need (Figure 1) These include
bull scientific and epidemiological understanding of SARS-CoV-2 transmission and infection
bull public health needs for combating the outbreak
bull technological capabilities of DCTT
bull performance of DCTT applications
bull ethical values and principles
bull characteristics of public adoption and acceptance and
bull legal issues and landscape
FIGURE 1 Interrelating Factors That Frame Responsible Development of Digital Contact
Tracing Technology
6 Digital Contact Tracing for Pandemic Response
The primary objectives for use of DCTT during the COVID-19 pandemic must be to reduce illness and death and facilitate public health efforts to reduce transmission of the virus These objectives fall under a broader overall goal of contributing to societal well-being during the pandemic It is not yet known whether and how much DCTT can contribute to these primary objectives nor whether it will be able to contribute without gen-erating new burdens or even harms such as incorrect warnings or ldquonoiserdquo that detract from the work of manual contact tracing
The process of identifying acceptable technology designs and uses is complex given the interplay among the factors Our analysis reveals that there is no ldquoone size fits allrdquo approach to DCTT There is variability across the United States with respect to SARS-CoV-2 prevalence and in-fection rates public health capacity public attitudes toward DCTT and acceptability of various potential features Moreover our understanding of SARS-CoV-2 and DCTT is evolving public health response needs and capabilities are changing and public attitudes are shifting Different tech-nologies used in different ways may be appropriate to achieve slightly dif-ferent public health goals in different localities and at different points in the pandemic A tiered and phased approach to technology development should be facilitated by law and policy prioritizing underlying interoper-ability while permitting user choices now and for the future
Given the complexity of the terrain as a first step those developing or considering widespread use of DCTT as part of pandemic response should be guided by the following principles and related actions (see box) These principles are meant to apply to DCTT as well as other dig-ital technologies used in novel ways during pandemic response
These principles make clear that in order to maximize the public good from use of DCTT public health needs and technological capabili-ties must be carefully aligned Government officials public health leaders leaders of other institutions employers digital technology developers and the public are all key stakeholders that must be informed and en-gaged in order to enable the most successful and ethically acceptable uses of DCTT
Guiding Principles for the Use of Digital Public Health Technologies for Pandemic Response
Transparency and public engagement are essential to an inclusive digital public health response
bull Government public health and digital technology leaders must engage
effectively with the public and other stakeholders to communicate the util-
ity importance oversight and limitations of relevant digital technologies
including their implications for individualsrsquo privacy and civil liberties
bull Transparency at all levels is essential for maintaining public trust and
confidence
bull To the extent possible digital public health responses should reflect the
range of values that are important to individuals including advancing the
health and well-being of the community as a whole
bull Decision makers should recognize the sacrifices that some people may be
willing to make during a pandemic in order to advance public health goals
Acceptance by some of particular monitoring capabilities should not be read
as a willingness to extend these methods to other problems or uses
Digital public health responses must represent the least infringement of civil liberties necessary to accomplish the public health goals
bull If preferred digital public health strategies infringe on privacy and other civil
liberties the infringements must be sufficiently justified by the circum-
stances of the pandemic offset by ample anticipated public benefit and
considered relative to infringements associated with other possible strate-
gies such as mass physical distancing
bull Only those data that are necessary and relevant for the stated public health
purposes should be collected Identifiable data should be stored in a secure
manner and only for the period of time that the public health purposes
require
bull Adopted technologies should not be used in ways that subject communities
to discrimination or surveillance for nonndashpublic health reasons
bull Respect for individual autonomy requires that users are sufficiently informed
of the public health goals of the technology and the extent to which those
goals are being met
Use of digital public health technologies and data must be guided by best available evidence
bull Decisions to deploy digital public health technologies should be based on a
careful assessment of the uses and limitations of any proposed technology
taking into account the best available evidence
bull Those who deploy digital public health technologies should continuously and
systematically monitor their performance as well as any evidence that is being
generated in other contexts about the selected technological solution and about
other competing technologies
bull Unintended consequencesmdashincluding those that might impact public health
goals core values and interests of the public and unfair advantage or disadvan-
tagemdashshould be carefully monitored and addressed as necessary
Responsible use of digital public health technology requires meaningful governance and accountability
bull Systems of governance must be trustworthy and well informed They must be
reviewed and adjusted as circumstances and evidence change or as unintended
effects are identified
bull Trusted representatives who are capable of developing and implementing uni-
form and fair standards for adopting and utilizing underlying digital technology
must be identified
bull Understandable transparent and publicly accessible rules must guide the
collection access control use storage and combination of data by government
authorities public and private institutions and other parties such as public
health researchers
bull Oversight accountability and consequences for abuse or misuse of these data
must be explicit and enforceable
The deployment of digital public health technology must be rooted in a commitment to equity
bull Digital public health technologies should be deployed in a manner that does
not propagate preexisting patterns of unfair disadvantage or further distribute
harms and risks unfairly throughout the population
bull To the extent possible digital public health technologies should be designed to
rectify existing inequities
bull Oversight mechanisms must be in place to ensure that the improved public
health outcomes are equitable and to detect and correct any unforeseen resul-
tant injustices attributable to the technology or that can be addressed using the
technology
bull The incentives and disincentives for adopting new technology must be equita-
ble not exploitative and aligned with effective use of the technology
bull Disparity-driven technology gaps should be explicitly recognized To the extent
possible provisions should be made to address the digital divide
Summary 9
Summary of Recommendations
The guidance document makes a number of recommendations related to (1) supporting effective and informed adoption of DCTT (2) designing flexible technologies to maximize public health utility while respecting other values (3) establishing meaningful processes for user disclosure and authorizationconsent (4) promoting equity and fairness in application of DCTT and (5) instituting transparent governance and oversight Here we provide a summary of recommendations
Supporting Effective and Informed Adoption
bull Those who authorize use of DCTT within a particular jurisdic-tion or institution should continuously and systematically monitor the technologyrsquos performance in that context This should include monitoring for effectiveness and benefit monitoring for harms and monitoring for the fair distribution of both benefits and harms They should also monitor evidence that is being generated in other contexts about their selected technological solution and about other competing technologies
bull Data collected through DCTT should be made available to public health professionals and to researchers in de-identified form to support population-level epidemiologic analysis
bull Data should be available to users that would permit them to further investigate their personal risk with public health officials or other health workers to add a layer of protection against unnecessary quarantine
bull Technologies or apps may produce some false negatives or false positives but they should be accurate enough that public health authorities feel confident that they support and donrsquot detract from contact tracing efforts
bull Trusted leaders should be enlisted to communicate effectively with the public about DCTT and encourage its use should the technol-ogy demonstrate some potential The limits of knowledge regard-ing effectiveness should also be explained along with what will be done to improve technological capabilities as understanding evolves
10 Digital Contact Tracing for Pandemic Response
bull Incentives can be a useful complement to encouragements how-ever any incentives for users to install and use DCTT must be equitable should not be coercive and should align with effective use of the technology
bull DCTT use should not be mandated at this time given uncertainty about potential harms and benefits Additional technology user and real-world testing is needed
Designing Flexible Technology to Maximize
Public Health Utility While Respecting Other Values
bull Technology companies should not alone control the terms condi-tions or capabilities of DCTT nor should they presume to know what is acceptable to members of the public
bull A ldquovalues in designrdquo approach to development of DCTT should be adopted (Flanagan Howe and Nissenbaum 2008 Knobel and Bowker 2011) Robust public and user engagement activities should be pursued to identify and incorporate to the extent pos-sible a range of values into the design of the technology These values may include privacy but also autonomy efficiency equity or others Technology design should reflect an appropriate balance and prioritization of identified values
bull Technology design should not be static but should be capable of evolving depending upon local conditions new evidence and changing preferences and priorities
bull DCTT should be designed to have a base set of features that pro-tect privacy with layers of additional capabilities that users may choose to activate An initial default should be that user location data are not shared but users should be provided with easy mech-anisms and prompts to allow for opting-in to this capability with encouragement to the public if it is shown to be critical to achieving public health goals
Establishing Meaningful Processes for
User Disclosure and Authorization (Consent)
bull A clear and concise module consisting of basic disclosure and vol-untary authorization should be developed to accompany DCTT
Summary 11
This module should not take the form of ldquoclickwraprdquo terms of service or end-user agreements but rather provide only essential information necessary for an individual to make a decision More detailed disclosures (such as FAQs in plain language) should be made easily accessible to those who wish to learn more with no hidden surprises
bull An opt-in approach to authorization should be instituted to ac-company initial DCTT rollout The feasibility and value of opt-out approaches should continue to be evaluated informed by what is technologically possible what local assessments of benefits and harms of the technology reveal over time and our evolving un-derstanding of the degree to which an opt-out approach is likely to increase or decrease utilization among different populations Opt-out approaches should not be precluded
Promoting Equity and Fairness in Application of DCTT
bull States localities and institutions that recommend widespread use of DCTT should provide technology (eg mobile phones Blue-tooth devices) and free data packages to those who desire but lack access to these devices
bull If there are lower rates of adoption of DCTT systems in some iden-tifiable communities public health authorities should find ways to compensate For example directing more non-DCTT resources and efforts toward those communities to meet specific needs that are elsewhere being supported by technology
bull If maps are generated based on DCTT to provide the public with the locations that COV+ individuals have visited steps must be taken to minimize the stigma and potential financial losses that could result from a location being identified as a hotspot
Instituting Transparent Governance and Oversight
bull Digital surveillance oversight committees should be established expeditiously with diverse and qualified membership to provide ethical and regulatory review prior to and concurrent with wide-spread use of a DCTT system
bull Understandable and publicly accessible rules must guide the col-
12 Digital Contact Tracing for Pandemic Response
lection access control use storage and combination of data by government authorities public and private institutions and other parties such as public health researchers
bull Only those data that are necessary and relevant for the public health response to COVID-19 should be collected and used
bull Identifiable data should be kept only for the period of time needed for the public health response to COVID-19
bull Identifiable data collected as part of this response should not be shared with anyone other than the relevant public health authorities without additional specific informed consent of individual users
bull Before a government or institution adopts a digital contact tracing program they should state the conditions under which the digital contact tracing program will be terminated
bull Future use of DCTT to advance public health or other efforts (eg use in seasonal flu surveillance) would require independent justification DCTT designed for public health use should not be used by law or immigration enforcement
bull The principles offered in this guidance document apply both during and following the COVID-19 pandemic
Legislative Recommendations
bull The United States Congress should enact new legislation specif-ically tailored to facilitate the use of DCTT as part of the public health response to COVID-19 while also protecting user privacy and ensuring data security
bull Congress should require DCTT developers to disclose to users in clear language the nature of the information that would be col-lected how it would be collected how it would be stored and for what purposes it may be used
bull While the rollout of DCTT should initially employ an opt-in au-thorization approach the feasibility acceptability and value of opt-out approaches should continue to be evaluated As such opt-out approaches to consent should not be precluded by legislation
Summary 13
bull Congress should prohibit the commercial use of data collected for COVID-19 response by DCTT
bull Congress should prohibit discrimination on the basis of data col-lected by DCTT
bull If Congress is unable to enact suitable legislation state legislatures should work toward enacting similar laws for their jurisdictions A ldquomodelrdquo state law should be rapidly developed to facilitate na-tionwide adoption of an appropriate law and uniformity of legal requirements
Summary of Analysis
Supporting Effective and Informed Adoption
The COVID-19 pandemic and the physical distancing efforts imple-mented to slow the rate of transmission have caused severe harm to indi-viduals communities and our society To protect the public good going forward we need a robust public health response that reduces the spread of SARS-CoV-2 and does so in a way that allows economic recovery to occur and to be sustained We also need to design and manage this public health response so as to minimize harms to individuals and society to distribute benefits and burdens equitably across the population and to avoid misuses of the technologies and the data they collect
To reduce the spread of SARS-CoV-2 chains of transmission need to be broken To do this people who have been exposed to SARS-CoV-2 or potentially exposed need to be identified as comprehensively and as quickly as possible so they can quarantine themselves and avoid infecting others This is the job of manual contact tracing by public health authori-ties in which people infected or presumptively infected with SARS-CoV-2 are interviewed and asked about their movements and interactions in-cluding where they work and shop how they travel with whom theyrsquove had contact and the nature of that contact (eg where the contact took place) Their contacts are then interviewed and potentially asked to quar-antine seek testing and take other protective measures if the contact is sufficiently high risk
14 Digital Contact Tracing for Pandemic Response
The hope is that DCTT can augment traditional contact tracing ef-forts either by working alongside and independently of manual contact tracing or by being integrated into manual contact tracing efforts in a way that makes these efforts faster more thorough and more efficient
Data suggest that a substantial proportion of transmissionsmdashper-haps as high as 50mdashoccur between individuals who are not symptom-atic and that transmission may occur as early as 3 days before onset of symptoms (WHO 2020) Because asymptomatic spread of SARS-CoV-2 appears to be a significant source of infection we need to identify po-tentially infected people before they show symptoms thus speed is of the essence This is one benefit of using DCTT potential contacts can be identified instantaneously notified quickly and asked to quarantine as soon as possible
Another benefit is identifying contacts who manual contact tracing methods may miss either because COV+ people do not remember all the places theyrsquove been or cannot identify all the people theyrsquove had contact with This is especially relevant given the long period of infectivity of SARS-CoV-2 which begins before people are symptomatic and aware they are infected (Ferretti et al 2020) If DCTT were designed to have optional location-monitoring capabilities this critical challenge could be mitigated even further For example location data might reveal that a COV+ person was at a restaurant at an exact time and date which could be followed up by contact tracers who could alert the public or use other measures to reach those who were also present in the restaurant at the same time In other disease contexts geolocation data have demonstrated some potential to support epidemiology and disease surveillance (see Fur-lanello et al 2002 Dredze et al 2013 Eckhoff and Tatem 2015 Fraccaro et al 2019) with technical cautions regarding accuracy and the like (Beu-kenhorst et al 2017)
One role for DCTT is to work alongside manual contact tracing but independently of it Individuals would download proximity tracing or exposure notification apps use them receive alerts if theyrsquove had a poten-tial contact with another user who is COV+ or presumptively COV+ and voluntarily self-quarantine without having contact with public health au-thorities or giving them data that feeds into public health contact tracing efforts It is possible that this would help to break chains of transmission and reduce the spread of SARS-CoV-2 though at this point these benefits
Summary 15
are speculative It is also possible that such exposure notifications will result in high rates of false positives
Another possible role is for DCTT to be integrated into manual con-tact tracing efforts When potential contacts are identified by DCTT they are connected to public health authorities who can then follow up with them There are different forms this could take and different kinds and amounts of data about contacts public health authorities could receive from DCTT On one end of the spectrum of reporting public health authorities would not receive individualsrsquo names or contact information only anonymous data The fullest version of reporting would securely send to public health authorities the names contact information such as address and phone number and other data about contacts that DCTT collected including data about their location and movement history
It is uncertain whether providing public health authorities with vol-umes of information on cases and contacts from DCTT will be useful in practice As mentioned above providing public health authorities with location data on cases and contacts collected by DCTT may help con-tact tracers to find and notify additional contacts However at present providing public health authorities with large amounts of data will be useful only if there is sufficient capacity to follow up on these data In addition there is a risk of low-quality data from DCTT flooding the system leading to investigation of false case contacts identified by DCTT and distracting from other important efforts Whether and to what extent data from DCTT will benefit contact tracing efforts is unknown pointing again to the importance of continuously collecting high-quality evidence about DCTT
Designing Flexible Technology to Maximize
Public Health Utility While Respecting Other Values
Use of DCTT is essentially an experiment as we have insufficient infor-mation about the performance of different DCTT and their efficacy In the face of this uncertainty how should DCTT be designed and how should its use be managed
Many efforts to advance DCTT in the United States and elsewhere have emphasized the importance of ldquoprivacy by designrdquo that is building privacy and security protections into the design of technology rather than counting on responsible use alone (Cavoukian 2010) As noted above
16 Digital Contact Tracing for Pandemic Response
some major technology companies have signaled this position through de-velopment of PPPT systems that embed features such as decentralization de-identified information user anonymity bans on collection of location data and minimal reliance on or integration of public health authorities or other government actors Many of these features have also been em-braced early by advocacy organizations (Crocker Opsahl and Cyphers 2020 Electronic Privacy Information Center 2020 Kahn Gilmor 2020) and in an open letter (ldquoJoint Statement on Contact Tracingrdquo 2020) from nearly 300 researchers These same actors have emphasized that use of DCTT should be fully voluntary
Although privacy is a key value individuals and communities may also value efficiency equity liberty autonomy economic well-being com-panionship patriotism or solidarity among other values People may accept more significant encroachments on privacy now if this ultimately results in realizing other values (such as companionship) that are of equal or greater importance to those individuals Rather than centering pri-vacy alone in design a different orientation is needed at this moment that of ldquovalues in designrdquo which incorporates a broader range of values into technology (Flanagan Howe and Nissenbaum 2008 Knobel and Bowker 2011) For example some users might wish to express autonomy solidarity or patriotism through DCTT by sharing their location history with public health professionals in order to advance the public health re-sponse increase system efficiencies (eg by contributing information that can lead to better data processing) and reduce the burden on essential workers At the same time there is value in further advancing autonomy by designing technology to allow individuals some control over what data about them are collected and shared
DCTT should be designed to have a base set of features that protect privacy and strive for interoperability but also should include other op-tional capabilities This could be achieved by designing DCTT to have a default that can be modified for example an initial setting could be that usersrsquo location data are not shared with public health authorities but us-ers may opt-in to this feature Such an opt-in approach is likely consistent with existing federal privacy laws
Designing DCTT this way gives users the flexibility to decide how to use the technology and how to engage with public health authori-ties consistent with their values and trade-offs they are willing to make This flexibility could also allow for more real-world evaluation of how
Summary 17
different users experience different features of DCTT in different loca-tions Technology design should not be static but it should be capable of evolving depending upon local conditions new evidence and changing preferences and priorities
DCTT developers must comply with a number of federal privacy laws These privacy laws generally permit the collection storage and use of personal information so long as the user provides meaningful consent Privacy law in the United States is generally sector-specific and limited in scope resulting in a patchwork of protections that differ significantly depending on the entity that collects the data and the type of data col-lected Given the complexity of existing federal privacy law and the need to further strengthen public trust in DCTT it would be beneficial for Congress to enact new privacy legislation that is specifically tailored to the use of DCTT in response to COVID-19 Such COVID-specific legisla-tion should be sensitive to the full range of values and recommendations described above
In short designing ldquomiddle-groundrdquo DCTT for flexible use may pro-vide the most adaptable and thus most robust public health responsemdashrespecting privacy and individual autonomy by allowing users to use DCTT in ways that express their own values
Public Acceptance of DCTT
While some groups have maintained that only PPPT-like minimal systems will be widely adopted because only they will earn and maintain public trust (Simpson and Conner 2020) there is insufficient evidence that pub-lic trust would be threatened by a DCTT system that has the capacity to securely collect location data integrate public health authorities and en-able voluntary sharing of certain user data (eg location data) with those authorities More research including through deliberative engagement sessions is needed to better understand how differences in the features and functionality of DCTT (such as optional sharing of geolocation data) influence trust and peoplersquos willingness to use DCTT Technology com-panies should not alone control the terms conditions and capabilities of DCTT nor should they presume to know what is acceptable to members of the public
Significant concerns have also been expressed by privacy advocates (Guariglia 2020) and in the popular press (Giglio 2020) about ldquosurveil-lance creeprdquomdashthat is a belief that state or corporate actors will use new
18 Digital Contact Tracing for Pandemic Response
surveillance technologies capacities and permissions well beyond the purposes for which they were initially justified to the public and beyond the time when they are useful for the COVID-19 pandemic Surveillance creep is a serious concern and should be carefully guarded against how-ever the possibility of surveillance creep is not a sufficient reason to limit development of DCTT to minimal systems Instead protections should be put in place to ensure that only those data that are necessary and relevant for the public health purposes at hand are collected and used and data should be kept only for the period of time needed for those pub-lic health purposes For this reason we would support COVID-specific legislation that would impose strict limits on the use of DCTT data for nonndashpublic health purposes
Finally the use of DCTT during the current pandemic should not set a precedent for future public health use (eg use in seasonal flu surveil-lance efforts) Future use would require independent justification Further use of DCTT in other contexts (eg by law enforcement or immigration enforcement) is presumptively unethical
Encouraging Adoption of DCTT
Researchers have estimated perhaps conservatively that DCTT use by 80 of smartphone owners or 56 of the population overall will be needed to suppress the epidemic (Hinch et al 2020) These estimates also highlight that some decrease in transmission would be realizable even with lower rates of technology adoption
In the United States many advocates and researchers have argued that use of DCTT must be fully voluntary However experience from other countries suggests that when use of a digital contact tracing app is voluntary only a minority of the population will download it Instead of making use fully voluntary and initiated by users there are ways that DCTT could be put into use without usersrsquo voluntary choice For exam-ple use of an app could be mandated as a precondition for returning to work or school or even further to control entry into a facility or trans-portation (such as airplanes) through scanning of a QR code to demon-strate personal exposure levels (Gan and Culver 2020)
While these approaches are hard to imagine in the United States some have argued that mandatory use of DCTT could be ethical If man-dates increase adoption of DCTT and improve the public health response
Summary 19
this would reduce the likelihood of lockdowns which are harmful and a severe limitation of individual liberty applied on a mass scale On the other hand mandated use of DCTT systems may not be effective People may not adhere to the mandate by simply leaving their phone at home Perhaps more important should the technology not deliver the hoped-for benefits having mandated the use of an unproven technology could result in a loss of public trust in the technology in the entity instituting the mandate and in the larger public health response potentially lead-ing to noncompliance with public health recommendations more broadly (Bernstein et al 2019)
Any decision maker considering mandatory use including govern-ment officials institutional leaders and employers must convincingly address a number of considerations Particularly important is the need to identify reliable evidence that the DCTT would be effective and to ensure that the burdens and benefits of use are equitable and justifiable At this time mandated use of DCTT by states or institutions is not jus-tifiable given uncertainty about potential harms and benefits Additional technology user and real-world testing is needed before mandatory use should be considered
As with any public health effort the amount of evidence that must be offered to illustrate that the intervention or program can achieve its aims and the degree to which people should be able to exercise choice in their participation should be in proportion to the anticipated bur-dens of the intervention or program For example the permissibility of mandating use of DCTT by the public depends on factors such as the sensitivity of the data that are collected the extent to which public health is integrated within the DCTT system and what actions are taken in response to confirmed virus exposure or being identified as COV+ (eg forced quarantine) The more burdens that are placed on individualsmdashfor example whether people are ordered into quarantine if they have been exposed to the virus or if there are limited social supports for those in quarantinemdashthe greater the demand should be on the performance of the DCTT system
Perhaps the most effective way to generate widespread US adoption of DCTT will be to offer incentives for its use in other contexts gener-ally speaking small incentives have been shown to lead to an increase in desired outcome (Singer and Ye 2013 Lee et al 2014) Given the impor-
20 Digital Contact Tracing for Pandemic Response
tance of widespread use modest incentives ought to be considered if and when there is sufficient evidence of the utility of DCTT so long as those incentives are not mandates in disguise Another ldquofirst linerdquo approach to increasing use of DCTT is for trusted community leaders public figures health care professionals and other respected individuals to communicate with the public and their communities about DCTT and to encourage its use through public engagement campaigns if and when the technology demonstrates sufficient potential
Establishing Meaningful Processes for
User Disclosure and Authorization (Consent)
Any effort to roll out DCTT should ensure that users have a meaning-ful opportunity to review and understand information about the specific technology and its uses Moreover given the importance of public trust and the current crisis of public trust in governments and technology com-panies handling private digital information there is a strong ethics argu-ment for requiring consent from individual users We recommend a care-fully crafted version of what is sometimes called simple consent which consists of basic disclosure and voluntary agreement or authorization This disclosure should include information about the purposes of the technology the userrsquos options for collecting and sharing data purposes for which data can be used and any known risks among other informa-tion This information should be presented in an accessible format on any DCTT app and more detailed disclosures should be readily accessible for those who wish to review them
Through an opt-in mechanism such as clicking a button to signal agreement users should be able to indicate their intention to use a DCTT The opt-in approach is consistent with mechanisms for agreement to use other downloaded applications An opt-in approach should be part of the initial introduction of DCTT given the novelty of the technology and its uses and the need to build trust and confidence in the system Successes of opt-out approaches in other areas suggest that the feasibility and value of an opt-out approach to DCTT should be carefully evaluated particularly in conjunction with assessment of whether public health goals are being met (Rithalia et al 2009) Such assessments should be informed by what is technologically possible by local data regarding benefits and harms of the technology and by evolving understanding of the degree to which
Summary 21
an opt-out approach is likely to increase or decrease utilization among different populations
Promoting Equity and Fairness in Application of DCTT
Digital contact tracing technology should be designed and used in ways that as far as possible promote an equitable distribution of benefits and burdens DCTT should be deployed in a manner that does not propagate preexisting patterns of unfair disadvantage or distribute harms and risks unfairly throughout the population It is well known that some commu-nities have lower rates of technology and data access and therefore may benefit less from use of DCTT unless steps are taken to address these digital disparities Additionally should use of DCTT be made a require-ment for entry into a workplace into a school or onto transportation then those who currently do not possess the required technology must not be unfairly burdened through lack of access In order to mitigate this states localities and institutions that recommend widespread use of DCTT should provide technology (eg mobile phones Bluetooth de-vices) and free data packages to members of the community who desire but lack access to these devices
Some populations may also experience greater harm and greater fear of harm from having their data collected For example some groups such as African Americans Hispanic Americans Muslim Americans and undocumented immigrants have more reasonable fear of their data being handed over to law or immigration enforcement and some groups have lower levels of trust in public health due to past injustices (Auxier et al 2019 CSM 2017 Pew Research Center 2017 Rodrigues et al 2018) This further substantiates the need to limit use of any data gathered by DCTT to its public health purpose
Instituting Transparent Governance and Oversight
DCTT must be developed with an eye toward both present and future implications We are rapidly gaining knowledge about SARS-CoV-2 and COVID-19 but still have essential gaps in our understanding In the United States public health responses including DCTT will generally be developed and coordinated by individual states regional consortia (Res-ton Sgueglia and Mossburg 2020) and associations Good governance in this context requires transparency and the creation of oversight bodies
22 Digital Contact Tracing for Pandemic Response
with the appropriate expertise and representation to allow nimble and effective responses while serving as trusted representatives
In order to address the range of ethics and governance concerns that relate to the design and use of DCTT we recommend that digital surveil-lance oversight committees be established perhaps at a state level and with a platform for national coordination These committees can provide ethics and regulatory review prior to and concurrent with widespread use of DCTT The committees should be composed of a diverse group of experts capable of evaluating the quality of a DCTT system locally including members of communities that experience higher rates of digital disparity
When assessing the design and use of digital contact tracing systems these committees (and the public more widely) should consider not only the risks and benefits accrued during the COVID-19 pandemic but also implications for the future How can we navigate safe use of these tech-nologies in a way that preserves public trust in them and enables the possibility of future beneficial use
As a start it should be emphasized that the principles offered in this and other guidance documents do not apply only during the pandemic Future efforts to advance DCTT capabilities during quieter times should make every effort to follow them
23
Introduction
Public health professionals around the world are working tirelessly to respond to the COVID-19 pandemic using tried-and-tested public health methods for infectious disease surveillance and control These traditional methods are essential to the global COVID-19 response To complement these actions and potentially augment the speed and efficacy of the public health workforce digital technologies are being harnessed Given the scale of the pandemic significant efforts are being undertaken to develop and leverage public-facing and health-system-supportive tech-nology solutions including smartphone apps and other digital tools that may aid public health surveillance and contact tracing
Digital contact tracing technology and closely related digital health products (hereafter DCTT) have been used in several countries as part of broader disease surveillance and containment strategies Globally many digital COVID-19 contact tracing strategies have already emerged in response to the pandemic This is not surprising given the ubiquity of mobile phones and other digital devices around the world (ldquoMobile Cellular Subscriptions (per 100 People)rdquo 2018) experiences developed during prior outbreaks and pandemics and the pre-COVID-19 momen-tum behind using digital technologies to support individual and health system capabilities (WHO 2017 Mathews et al 2019 Aiello Renson and Zivich 2020 Mahmood et al 2020) In the United States DCTT has been proposed as an integral part of some plans to ldquoreopenrdquo the country (Allen et al 2020 Hart et al 2020 Simpson and Conner 2020) It is al-most certain that these and related technologies will become part of not only the COVID-19 response but also the larger toolbox for future public health communicable disease prevention and control
24 Digital Contact Tracing for Pandemic Response
While novel public health surveillance technologies such as DCTT have theoretical promise their effectiveness is unclear These technologies also raise important ethical legal and governance challenges that require comprehensive analysis in order to support decision-making regarding their appropriate use A number of frameworks recommendations and analyses have emerged recently in an effort to chart potentially ldquosaferdquo pathways for use of public health disease surveillance technology Many in the United States such as the Electronic Frontier Foundation Elec-tronic Privacy Information Center American Civil Liberties Union and the Center for American Progress are proposing that digital public health surveillance technologies must embrace strict data privacy protections decentralized data storage a high degree of anonymity and voluntary adoption (Crocker Opsahl and Cyphers 2020 Electronic Privacy Infor-mation Center 2020 Kahn Gilmor 2020 Simpson and Conner 2020) Others have argued that technologies that seek to enhance public health response during a pandemic should more closely align with the needs of public health professionals and the evidence-based procedures they follow stating that interests in serving the publicrsquos health ought to weigh more heavily in the necessary balancing of stakeholder interests (de Jong et al 2019 Watson et al 2020) This view is in part based on a recogni-tion that during countless other outbreaks the public has benefited from traditional disease surveillance and contact tracing which are heavily re-liant on centralized data storage and when necessary the collection of identifiable information These traditional approaches are governed by ethics principles (PHLS 2002) ethics guidelines (WHO 2017) and laws (ASTHO 2012) and digital technologies represent a new tool to support them
While debates and recommendations about appropriate design and use of DCTT have focused intensely on minimizing important data-related risks a wider lens is needed to fully appreciate the many additional criti-cal questions that need attention This report begins to grapple with these questions which are critical to address in order to guide responsible use of DCTT Given the complexity of the terrain as a first step toward estab-lishing a foundation for responsible decision-making regarding potential use of DCTT we offer a set of guiding principles (see box) These prin-ciples are meant to apply to DCTT as well as other digital technologies used in novel ways during pandemic response
Guiding Principles for the Use of Digital Public Health Technologies for Pandemic Response
Transparency and public engagement are essential to an inclusive digital public health response
bull Government public health and digital technology leaders must engage
effectively with the public and other stakeholders to communicate the util-
ity importance oversight and limitations of relevant digital technologies
including their implications for individualsrsquo privacy and civil liberties
bull Transparency at all levels is essential for maintaining public trust and
confidence
bull To the extent possible digital public health responses should reflect the
range of values that are important to individuals including advancing the
health and well-being of the community as a whole
bull Decision makers should recognize the sacrifices that some people may be
willing to make during a pandemic in order to advance public health goals
Acceptance by some of particular monitoring capabilities should not be read
as a willingness to extend these methods to other problems or uses
Digital public health responses must represent the least infringement of civil liberties necessary to accomplish the public health goals
bull If preferred digital public health strategies infringe on privacy and other civil
liberties the infringements must be sufficiently justified by the circum-
stances of the pandemic offset by ample anticipated public benefit and
considered relative to infringements associated with other possible strate-
gies such as mass physical distancing
bull Only those data that are necessary and relevant for the stated public health
purposes should be collected Identifiable data should be stored in a secure
manner and only for the period of time that the public health purposes
require
bull Adopted technologies should not be used in ways that subject communities
to discrimination or surveillance for nonndashpublic health reasons
bull Respect for individual autonomy requires that users are sufficiently informed
of the public health goals of the technology and the extent to which those
goals are being met
Use of digital public health technologies and data must be guided by best available evidence
bull Decisions to deploy digital public health technologies should be based on a
careful assessment of the uses and limitations of any proposed technology
taking into account the best available evidence
bull Those who deploy digital public health technologies should continuously and
systematically monitor their performance as well as any evidence that is being
generated in other contexts about the selected technological solution and about
other competing technologies
bull Unintended consequencesmdashincluding those that might impact public health
goals core values and interests of the public and unfair advantage or disadvan-
tagemdashshould be carefully monitored and addressed as necessary
Responsible use of digital public health technology requires meaningful governance and accountability
bull Systems of governance must be trustworthy and well informed They must be
reviewed and adjusted as circumstances and evidence change or as unintended
effects are identified
bull Trusted representatives who are capable of developing and implementing uni-
form and fair standards for adopting and utilizing underlying digital technology
must be identified
bull Understandable transparent and publicly accessible rules must guide the
collection access control use storage and combination of data by government
authorities public and private institutions and other parties such as public
health researchers
bull Oversight accountability and consequences for abuse or misuse of these data
must be explicit and enforceable
The deployment of digital public health technology must be rooted in a commitment to equity
bull Digital public health technologies should be deployed in a manner that does
not propagate preexisting patterns of unfair disadvantage or further distribute
harms and risks unfairly throughout the population
bull To the extent possible digital public health technologies should be designed to
rectify existing inequities
bull Oversight mechanisms must be in place to ensure that the improved public
health outcomes are equitable and to detect and correct any unforeseen resul-
tant injustices attributable to the technology or that can be addressed using the
technology
bull The incentives and disincentives for adopting new technology must be equita-
ble not exploitative and aligned with effective use of the technology
bull Disparity-driven technology gaps should be explicitly recognized To the extent
possible provisions should be made to address the digital divide
Introduction 27
In reflecting on these principles it becomes clear that if we wish not only to realize but to maximize the public good that might come from use of DCTT we must carefully define and responsibly align public health needs and capabilities with technological needs and capabilities We must understand that although technology may serve as a workforce multiplier it alone will not solve the public health challenges we face We must identify and address assumptions and misinformation about technologies and data use We must provide the means and opportu-nity for informed decision-making by the public and those who serve as our representatives Government officials public health leaders leaders of other institutions employers digital technology developers and the public all must be adequately informed and engaged in order to make the best decisions possible under the circumstances
ONE
29
Types of Information Collected through Contact Tracing
Data Collected from Infected Persons
Symptoms and Course of Illness
Information about COVID-19 patientsrsquo signs symptoms and course of illness is important to public health because it provides a basis for refining clinical case definitions and informing health care providers and the gen-eral public (CDC 2020c) This includes the specific signs and symptoms manifested by persons who are COV+ as well as the relative frequency and durations of different signs and symptoms This would also take into consideration those persons with no symptoms but who test positivemdashthose who are presymptomatic (develop symptoms later) those who are postsymptomatic (clinically recovered but still infectious) and those who never manifest illness at all
Typically contact tracing begins with a case in which a person has confirmation of infection by means of a diagnostic test However in some cases test results are not reported until several days later and individuals may be identified as ldquopresumptive positiverdquo cases until testing can be completed In these cases contact tracing efforts will need to be updated when test results are returned For example if a test comes back negative public health professionals will want to notify contacts that they no lon-ger need to quarantine
Public Health Perspective
30 Digital Contact Tracing for Pandemic Response
Movement and Contacts
In order to manage cases appropriately (identify and track the infected isolate the sick quarantine the exposed) public health officials need de-tails on each case (Resolve to Save Lives nd) First they need to know who and where the individual is That means personally identifiable infor-mation and contact information (address phone numbers email) It also means information about the nature intensity and duration of contact with individuals to whom they may have transmitted the disease This may include information about where the individual works and the kind of work they do (eg health care worker) how they travel (eg bus sub-way car) and where they shop or any other public venues they may have visited during a period of possible infectiousness (PIH 2020a) It may be helpful in certain circumstances for public health officials to ensure that suspected cases contacts or other high-risk individuals are following iso-lation and quarantine recommendations or orders
Contact tracing involves identifying all individuals who have had sig-nificant exposure to confirmed or probable cases during the time prior to and after the onset of symptoms both of which are times when the case is thought to be infectious (Africa CDC 2020) Contacts could be those who are caring for COVID-19 patients especially if they lacked proper PPE and those who had close interaction with the COV+ person over a sustained period of time particularly in enclosed spaces (PIH 2020a) For COVID-19 contacts are identified by asking a person with a confirmed or probable case about people they may have been within 6 feet of for 15 minutes or more starting from 48 hours before the onset of symptoms and lasting until the person is isolated (CDC 2020b)
Data Collected from Contacts of Infected Persons
Contact Details
In addition to the data collected from individuals with COVID-19 con-tact tracers will collect data from potentially exposed individuals (con-tacts) Information about the nature intensity and duration of contact with an infected person may be collected for a contact if information about the case is known to the contact These details can help a contact tracer more accurately determine whether the contact is at high or low risk for SARS-CoV-2 transmission and help determine whether a con-tact should quarantine for 14 days (the upper bound of the SARS-CoV-2
Public Health Perspective 31
incubation period) In addition public health professionals may gather contactsrsquo demographic information and other personal data to contribute to population-level disease surveillance and situational awareness about an epidemic (CDC 2005) However the information needed at baseline is only a personrsquos name and contact information
Symptoms (If They Develop) and Course of Illness as well as Information about Close Contacts
If a contact develops COVID-19 symptoms while in quarantine andor tests positive for the virus public health will then collect the data required for a COVID-19 case This includes collecting information on the con-tacts that a person may have had (if any) in the days immediately before and during the course of their infection
How Contact Tracing Information Informs Public Health Action
To reduce disease burden and help make ldquoreopeningrdquo safer during the COVID-19 pandemic the United States and other countries will need to identify gather information about and safely isolate cases and quar-antine their contacts to reduce community transmission (Watson et al 2020) Gathering information about possible cases and their contacts en-ables public health to break chains of transmission
Contact tracing involves stages (CDC 2020a) including
1 identifying an infected person as a COVID-19 case
2 identifying the close contacts of that case (Africa CDC 2020)
3 getting in touch with contacts
4 asking contacts to quarantine at home for 14 days
5 assessing contacts for possible symptoms and
6 following up with COV+ persons and their contacts to identify new or worsening symptoms and connect them with medical care if needed
Contact tracers also play an important role in providing resources for COV+ persons who are in home isolation and their contacts who are
32 Digital Contact Tracing for Pandemic Response
in home quarantine Knowing who and where cases and contacts are can enable provision of supplies such as digital thermometers or masks Effective contact tracing that enables isolated cases and quarantined con-tacts to remain at home also requires providing a range of social sup-port services or ldquocare packagesrdquo from delivering food and medicines to trash pickup Furthermore vulnerable individuals who are homeless or otherwise unable to sufficiently isolate or quarantine in their current living conditions may need to have alternative housing arranged to safely remain separated from others (CDC 2020b)
Finally contact tracers explain what quarantined contacts should do if they begin to develop symptoms consistent with COVID-19 (Africa CDC 2020) Depending on the context contact tracers may engage in active monitoring by regularly communicating with contacts about their health status through phone text message or possibly mobile applica-tions In rare cases public health can make quarantine mandatory and may monitor a quarantined individual to ensure that they do not break quarantine Contact tracers may also facilitate access to health care by providing telemedicine resources or other information and support for accessing medical care
Characteristics That Make Data Useful to Public Health for Reducing Disease Transmission
Data Access
If digital contact tracing technology and closely related digital health products (together DCTT) are intended to support the public health ac-tions described above and directly amplify public health capacity to con-duct case identification and contact tracing then data collected through DCTT must be accessible to public health authorities Identifying infor-mation and location data for cases and contacts of cases are necessary for public health use so that contact tracers can do their work to uncover ongoing transmission and enable isolation and quarantine These data should also be durable meaning that public health can return to the data in order to interact with and support cases and contacts These data can also be useful at a population level if de-identified and aggregated by illuminating trends in community transmission and providing support for decisions about resource allocation
Public Health Perspective 33
Data Format
Data should be provided to public health authorities in a usable format that is compatible with public health systems and that has the granularity and specificity of personal information that is needed for use in contact tracing Without personal identifiers the data cannot be used by public health workers to undertake contact tracing Data should also contain information about the nature of a contact including the proximity of the contact and number of minutes that the person was in contact with an infected individual Location data can also help public health author-ities to conduct contact tracing particularly when contact occurred in a crowded area and involved people who donrsquot know one another Loca-tion data from a case can help public health professionals identify con-tacts even when those contacts themselves are not using a contact tracing app because the data shows contact tracers where to look for additional contacts
Data Accuracy
Data that identifies individuals as having sustained contact with a case must be as accurate as possible If criteria for being considered a contact are too restrictive it may result in missed contacts and sustained chains of disease transmission If criteria are too broad it may result in unneces-sary restriction of movement which could have significant personal and economic consequences
Timeliness of Data
Data from cases and contacts must be timely in order to enable case-based management that will help reduce community transmission For contact tracing to be effective infected individuals need to be isolated and their contacts identified and quarantined as quickly as possible Testing for SARS-CoV-2 can take time sometimes many days for a test result Especially because SARS-CoV-2 is transmissible during the pre-symptomatic period data on symptomatic individuals should be made available to public health officials even before a positive test is returned in order to enable identification and quarantine of contacts right away If this information is delayed until a test result is received it may be too late to identify and quarantine contacts because contacts (if infected) will already be contagious and may have spread the virus to others
34 Digital Contact Tracing for Pandemic Response
VolumeAvailability of Data
The more that individuals opt to share their information to support con-tact tracing the more effective contact tracing will be in breaking chains of viral transmission and controlling epidemics of COVID-19 The exact proportion of cases and contacts that need to be identified in order to avoid large surges of cases which overwhelm health care systems is un-certain but the goal is to identify all infected cases and all close contacts of each case (PIH 2020b)
Recommendations
bull Technologies or apps with the goal of enhancing public health ca-pacity to identify cases and trace contacts in order to control the spread of SARS-CoV-2 should be designed to match functionality with that goal
bull Technologies or apps may produce some false negatives or false positives but they should be accurate enough that public health authorities feel confident that they support and donrsquot detract from contact tracing efforts
bull DCTT approaches for public health should be designed to facilitate the following
deg identifying contacts including those who may not be easily found otherwise
deg finding and notifying contacts rapidly before they develop symptoms if infected
deg analyzing the nature of contact to determine whether contact is high medium or low risk and to support decisions about whether a contact should quarantine and
deg following up with cases and contacts so that public health can provide resources to support isolation and quarantine
bull Data collected through DCTT should be made available to public health professionals and to researchers in de-identified form to sup-port population-level epidemiologic analysis
T WO
35
Characteristics of SARS-CoV-2 Relevant to Candidate Digital Solutions
The SARS-CoV-2 virus has some unique transmission characteristics and clinical manifestations that can help guide use of digital contact trac-ing solutions Individuals infected with this virus may or may not show symptoms or may show a range of different and sometimes nonspecific symptoms Estimates regarding the percentage of individuals who are in-fected but never develop symptoms is highly uncertain ranging from 5 to 50 (Heneghan Brassey and Jefferson 2020) Data suggest that a substantial proportion of transmissionsmdashperhaps as high as 50mdashoccur between individuals who are not symptomatic and that transmissibility may extend out as long as 3 days before the onset of symptoms (WHO 2020)
The complexity of asymptomatic and presymptomatic transmission makes it more difficult to identify all cases of COVID-19 It also means that manual contact tracing is less effective because people are unlikely to remember all of their contacts during the long period of infectivity (Ferretti et al 2020) however it does not negate the need for contact tracing Identifying symptomatic cases will still help greatly with slowing the spread because their contacts can be asked to quarantine to prevent them from spreading the virus if they are indeed infected This means that whether they become symptomatic or not contacts will be quarantined and the chain of transmission will be broken If contact tracing can be implemented on a large enough scale perhaps with support from DCTT
Digital Technology and Contact Tracing
36 Digital Contact Tracing for Pandemic Response
eventually the virus could be managed at much lower levels of community transmission and large epidemics of unrecognized spread will not occur
The transmissibility of the virus when a person has no symptoms further suggests that effective solutions may require multimodal inter-ventions combining contact tracing with frequent rapid and ubiquitous testing and continued social distancing to varying extents (Cheng et al 2020)
Because of presymptomatic spread contact tracing efforts and dig-ital solutions to augment those efforts should support identification of contacts a person had 2 days before their symptoms and at least 3 days after the resolution of those symptoms (if the person continued to have contacts through that time period) (CDC 2020d) Additionally public health messages delivered by these technologies should urge contacts to quarantine for the full 14-day incubation period
Previously Existing Contact Tracing Technologies
Prior to this pandemic health agencies in high- medium- and low-income countries had begun to develop and use digital tools to augment the man-agement of infectious diseases including sexually transmitted infections (HIV chlamydia gonorrhea) and high-consequence epidemics (Ebola) (Danquah et al 2019)) However these have been primarily used to facil-itate case interviews partner notification (in the case of STIs) and record keeping as opposed to fully digitizing or automating the contact tracing process
It has been recently suggested that digital contact tracing could con-tribute to the management of the ongoing COVID-19 pandemic and the experiences of containing SARS-CoV-2 in countries such as China Sin-gapore and South Korea provide noteworthy examples However un-dertaking this case-based intervention on the scale required to achieve pandemic control is a novelty in the history of public health Although technological development is proceeding rapidly several foundational is-sues have yet to be resolved including functionality connectivity to pub-lic health authorities and informatics systems usability by disease inter-vention specialists (DIS also referred to as contact tracers) and sufficient protection of personally identifiable information among others
Digital Technology and Contact Tracing 37
Introduction of Novel Digital Contact Tracing Technologies
Digital contact tracing technologies and platforms have recently been in-troduced and the CDC has published preliminary criteria for evaluating these tools (CDC 2020e) It can be helpful to consider three broad ap-proaches along the spectrum of potential methods of digital contact trac-ing a maximal approach (typified by the South Korean governmentrsquos cen-tralized data collection (M S Kim 2020)) a minimal approach (typified by the AppleGoogle decentralized privacy-protecting proximity tracking (Apple and Google nd)) and a diverse middle ground that aims to aug-ment manual contact tracing with the collection of digital data Perhaps the most promising approach in this middle ground involves allowing us-ers to turn over both proximity data and GPS location data (ie cell-site location data) to public health authorities on a voluntary basis
Along with this ldquominimal to maximalrdquo spectrum in the design of dig-ital contact tracing technologies and systems there is another spectrum that concerns voluntary versus mandatory use of these technologies are individuals entirely free to use these technologies or not or should poli-cies incentivize or even mandate their use At one extreme South Korea (Republic of Korea) implemented a system (called Safe Korea) supported by the Ministry of the Interior and Safety that collects a variety of per-sonal data in a centralized database in order to enforce quarantine orders and track possible contacts (M S Kim 2020) Israel also implemented a centralized involuntary data collection system for tracking COVID-19 cases and alerting those who may have been exposed (Hendrix and Eg-lash 2020) In Poland health authorities have set up mandatory ldquocheck-insrdquo involving a GPS-waypoint capture and ldquoselfierdquo photographs sent to the monitoring agency to ensure that individuals are not breaking quar-antine (Hamilton 2020)
These centralized systems can be designed to incorporate data from a variety of sources The data collected include location data from mobile phones QR codes can also be scanned to track the use of public transit where GPS data may be inadequate (due to low resolution) to accurately distinguish the occupants of one vehicle from another The data collected from mobile phones can then be integrated with data from other sources such as facial-recognition cameras credit card transactions and social media
38 Digital Contact Tracing for Pandemic Response
At the other extreme of technology invasiveness for contact trac-ing isolation and quarantine many corporations and working groups (including the AppleGoogle collaboration) have developed privacy-pre-serving proximity tracking (PPPT) using Bluetooth Low Energy (BLE) ldquohandshakesrdquo that record close contact between mobile phone users In most architectures these proximity data are stored in the usersrsquo phones as anonymized ldquobeaconsrdquo that cannot be used to re-identify the users directly If a user with a PPPT app installed on their phone tests positive and enters test results into their app those who have been identified as having been in close proximity to them can be notified by the app This notification can be automatic or at the discretion of the person who is COV+ depending on the app design If notified a user who has been in contact with a COV+ individual would receive a push notification alert-ing them to possible exposure (which may be timestamped) but with no other identifying information
Because of its reliance on anonymized data PPPT on its own is dis-tinct from manual contact tracing In recognition of this fact some de-signers and researchers now use the more descriptive term ldquoexposure no-tificationrdquo Moreover the public health usefulness of PPPT is uncertain it is unclear how PPPT can best be used in tandem with manual contact tracing especially if the data it collects are inaccessible to or unusable by public health authorities It remains to be seen whether PPPT will provide significant benefit operating alongside but not integrated into manual contact tracing
Between these extremes there are a number of possible middle-ground approaches that aim to strike a balance among public health utility tech-nological feasibility and user privacy protections This middle ground divides into two rough categories centralized storage of de-identified data and decentralized storage of personally identifying data The United Kingdomrsquos NHSX is reportedly developing an app that would utilize BLE handshakes to collect anonymized proximity data which would then be stored on a centralized government-operated server
The most prevalent middle-ground approach in the United States context involves the collection and storage of personal datamdashincluding identifying information and location datamdashon the userrsquos phone This de-centralized but personally identifiable data can then be voluntarily shared with public health officials if the user tests positive for SARS-CoV-2 For
Digital Technology and Contact Tracing 39
example an MIT team has developed an app called Private Kit Safe Paths (MIT nd) that stores usersrsquo location data on their phone for 28 days If a user tests positive she can voluntarily upload her location data to a website that is accessible only to public health officials Officials can then analyze this personally identifiable data and subsequently broadcast re-dacted and de-identified data to other users Healthy users would have access to these redacted location data of COV+ users but their own data would not leave their phones (The developers plan to incorporate BLE proximity data once available) Along similar lines the North Dakota state government has rolled out an app that stores both location data and proximity data on a userrsquos phone which can be voluntarily released by the user to public health authorities if the user tests positive (NDDoH 2020) At a minimum the storage of user location data can function as a ldquomemory aidrdquo if the user tests positive But releasing the data to public health authorities may help them analyze the spread of COVID-19 and alert individuals or groups that have been in contact with persons who are COV+ An overview of various DCTT apps and platforms as well as features that are relevant to this analysis are provided in Table 1
Because DCTTs are so new very little is known about their actual utility to public health authorities for controlling this pandemic Although multiple countries that have had success in greatly reducing transmission of SARS-CoV-2 have included DCTT in their response these countries have employed multiple simultaneous approaches to controlling the vi-rus including manual contact tracing and it is difficult to disentangle what made those responses successful Preliminary impressions from Ice-land may suggest that DCTT at least in that context had a small impact on reducing transmission ldquoespecially compared with methods of manual contact tracing such as phone callsrdquo (Hadavas 2020) This is with the highest public download rate of any DCTT app thus far
DCTTs have the potential to be helpful but they also have the poten-tial to distract from other public health efforts including manual contact tracing Concerns about implementation of DCTT from the public health perspective include that data generated may not be useful to public health authorities either because they donrsquot include detailed data to aid contact tracers or because the data are inaccurate (Mills Rodrigo 2020) DCTT if not calibrated well could be overly inclusive and create many false positives This would be harmful to those individuals being notified and
Purp
ose
Tech
nolo
gies
Use
dD
ata
Stor
age
Part
icip
atio
n
Proximity-based exposure notification
Digital contact tracing (DCT)
Bluetooth LE
GPS
SMS
Centralized
Decentralized
Mandator (actually or functionally)
Voluntaryopt-in
Gov
ernm
ent a
cces
sA
pp N
ame
Dev
elop
er o
r Cou
ntry
Max
WeC
hat
Alip
ayCh
ina
D
ata
com
es fr
om g
over
nmen
t so
urce
s lo
catio
n da
ta s
ent t
o po
lice
Intervention Type
Middle Ground
Trac
e To
geth
erSi
ngap
ore
M
anda
tory
gov
ernm
ent
acce
ss if
pos
itive
NH
SXO
xfor
dO
xfor
d
Gov
ernm
ent m
aint
ains
dat
a
Nex
tTra
ceFr
ed H
utch
inso
n Ca
ncer
Re
sear
ch C
ente
r
Gov
ernm
ent m
aint
ains
dat
a
but n
o st
orag
e
COVI
D
Safe
Path
sM
IT
Volu
ntar
y up
load
by
user
s
who
test
pos
itive
Aar
ogya
Set
uIn
dia
A
nony
miz
ed a
ggre
gate
Care
19N
orth
Dak
ota
In
agg
rega
te o
ptio
nal i
f pos
itive
Minimal
Covi
dSaf
eU
niv
of W
ashi
ngto
n
Non
e
Covi
dWat
chU
niv
of S
tanf
ord
amp U
niv
of W
ater
loo
To
val
idat
e te
st re
sults
CoEp
iCo
Epi
O
pt-in
to s
hare
BT
and
sym
ptom
lo
g w
ith C
oEpi
ser
ver
itoG
erm
any
N
one
pos
itive
resu
lts to
ito
serv
er
TAB
LE 1
Ex
ampl
es o
f Dig
ital C
onta
ct T
raci
ng T
echn
olog
ies
to S
uppo
rt A
ctiv
e Pu
blic
Hea
lth S
urve
illan
ce a
nd R
elev
ant F
eatu
res
Digital Technology and Contact Tracing 41
asked to quarantine unnecessarily and it could result in large proportions of the population remaining at home at any one time Individuals living or working in congregate settings could receive frequent notifications that would result in their inability to leave quarantine for long periods of time Finally public health authorities could also become inundated by data from these technologies and not have sufficient approaches to manage or analyze the incoming information
Relevant Differences between Manual and Digital Contact Tracing
There are several noteworthy differences between manual contact tracing efforts and use of DCTT First there is a significant amount of evidence regarding the effectiveness of manual contact tracing which is lacking for DCTT Second manual contact tracers interact with individuals who are confirmed or suspected cases and contacts of cases but not other members of the general public DCTT intervention would affect all users regardless of circumstances (though some more than others) Third manual contact tracing occurs most often through human-to-human encounters with the opportunity to clarify misconceptions address worries and express sympathy and other important affects DCTT can certainly incorporate sharing of important information and potentially communicate some af-fect but it currently lacks a range of other human capabilities and char-acteristics Fourth there typically are fewer data intermediaries in manual contact tracing (fewer entities handling data) in DCTT a valid argument could be made that a wide range of technology developers (and perhaps mobile network operators) must remain connected to relevant data in order to continuously identify problems and improve functionality
It is because of these and other differences that DCTT has been pro-posed as a potential complement to rather than a replacement for man-ual contact tracing However over time it is possible that technology could develop to close gaps between some of these differences (if and as needed) and in parallel the goals of contact tracing and public health surveillance may evolve
Ethics of Designing and Using DCTT
43
THREE
43
Those developing DCTT and those considering its use should systemat-ically take into account and document alignment with the guiding princi-ples outlined in this report
When considering the ethics of DCTT key ethical questions con-cern the features that DCTT should have (eg should digital contact tracing apps collect usersrsquo location data) whether and how individualsrsquo data should be shared with public health authorities how ethically to encourage use of DCTT (eg under what circumstances would it be eth-ical to incentivize or mandate use of DCTT) what kind of supports and equity-promoting measures should accompany use of DCTT and how governance and oversight of DCTT should be structured
The sections that follow consider these questions one by one A key conclusion of this report is that these features of the design and use of DCTT are ethically interrelatedmdashreaching a determination regarding any one question requires careful consideration of them all Rather than reaching ldquoone size fits allrdquo conclusions about specific features of uses of DCTT decision makers should ethically assess DCTT systems holistically
Generally a public health measure is ethically justifiable if it strikes a reasonable balance between competing considerations and if it pro-vides sufficient public health benefit (or the prospect of benefit) to justify the burdens associated with it DCTT systems are ethically justifiable if they strike a reasonable balance between multiple ethical considerations including
bull enabling an effective and efficient public health response
bull protecting individual privacy and preventing harms to individuals
44 Digital Contact Tracing for Pandemic Response
including harms from sensitive data being revealed and from erro-neously being subjected to isolation or quarantine orders
bull allowing individuals to control what information about them is collected and revealed to whom including through appropriate dis-closure and authorization processes for data collection
bull promoting equitable distribution of benefits and burdens of DCTT
bull maintaining public trust in DCTT and in the COVID-19 public health response and
bull taking seriously the future implications of decisions that we make today
To illustrate a holistic assessment consider whether it is ethically jus-tifiable for an employer to mandate that employees use a DCTT as a condition of returning to work This will depend upon many features of the DCTT system what kind of data the DCTT collects (eg does it collect location data or just record proximity events) whether there is public health capacity to make good use of these data what the data are used for (eg will the employer ban an employee from the workplace on the basis of a DCTT-identified contact) what kind of social supports are available (eg is there paid leave for employees) what employeesrsquo attitudes are toward use of DCTT and whether mandating use is likely to have public health benefit among other factors These factors may vary from place to place and may change over the course of the pandemic Thus there is no ldquoone size fits allrdquo ethically optimal approach to DCTT
Justifying the Use of DCTT Systems
A foundational issue is why deploying any DCTT during a pandemic is justified given there are manual contact tracing capabilities that are well established while the performance and effectiveness of novel technolo-gies is less established The need to move quickly to minimize the spread of the virus poses challenges here as the data needed to fully make the case that these technologies substantially contribute to the public health response may not be available prior to widespread use The primary ar-gument for DCTT is that the capacity of manual contact tracing may be
Ethics of Designing and Using DCTT 45
exceeded and we may not be able to bolster the public health workforce rapidly and sufficiently enough to meet needs DCTT has the potential to quickly and exponentially expand the reach of contact tracing In ad-dition DCTT may allow more efficient identification and quarantine of potential contacts of COV+ people than manual contact tracing alone particularly given the high number of infections that have been spread by asymptomatic individuals
Nonetheless reasonable people disagree about the prudence of pur-suing DCTT especially given its limited performance history and poten-tial risks including diverting attention and resources from more effective interventions The limited attention and resources available during a pan-demic must be allocated efficiently and effectively
To justify potentially widespread use of technologies such as DCTT therefore a number of considerations must be addressed
bull whether the technology is designed to meet an important and unmet public health need
bull whether there is sufficient evidence or reason to suggest that the technology will be effective at serving its purpose
bull whether the outbreak is characterized by sufficiently severe morbid-ity and mortality and a high rate of disease transmission to warrant large-scale introduction of novel systems
bull whether there are other less autonomy-restricting or less risky al-ternatives to widespread use and
bull whether it is reasonably likely that a sufficient number of individ-uals will use the technology to achieve the intended public health benefit
Monitoring and Evaluating Technologies to Inform Policy and Practice
A number of public health ethics principles necessitate the ongoing mon-itoring and evaluation of DCTT systems First DCTT must be shown to perform reasonably well at achieving its stated goal reducing the spread of SARS-CoV-2 The effectiveness of DCTT programs should be illus-trated at a number of stages
46 Digital Contact Tracing for Pandemic Response
1 Robust initial technology testing is needed to publicly justify the widespread adoption of DCTT and avoid public failures which may hamper future uptake (eg Lovejoy 2020 Morse 2020) This typically includes alpha testing in virtual environments and beta testing in different community settings
2 If and when a DCTT is implemented on a wide scale it must be monitored on an ongoing basis to assess reach effectiveness func-tionality best practices and any harms
3 When approaching a previously identified stopping point for use of DCTT monitoring can help to identify when utilization is no longer needed
If at any of these points evidence clearly suggests harm (particularly in comparison to other methods that the public might find more acceptable) this evidence should provide a basis upon which to revisit strategies pri-orities and allocation of resources Attention should be given to foresee-able side effects that may dramatically influence the overall effectiveness of the program such as individuals carrying their smartphones around with them selectively so as to avoid particular undesired consequences of DCTT policies
Anonymized aggregate data including user feedback must be eval-uated to ensure that benefits and burdens are distributed fairly As noted earlier unintended burdens may include inequitable outcomes that may arise in a DCTT program for example resulting from uneven access to the required technology to participate disparate concerns about sur-veillance within some communities that might limit widespread use or discrimination that may result from being identified as COV+ due to the program or for communities that are termed ldquohotspotsrdquo based on maps of COV+ location data Additionally it is possible that some communi-ties might get higher rates of false positives because they are located in densely populated areas thus increasing the burden of self-isolation If any of these inequities are identified steps must be taken to mitigate them
Finally numerous actors should engage in the monitoring and eval-uation of DCTT systems Technology developers and public health re-searchers have a clear role in this process Technology developers should work with public health researchers to monitor accuracy precision func-
Ethics of Designing and Using DCTT 47
tionality confidence of estimates sources of error and the like Research-ers may also be able to contribute innovative methods to systematically and rapidly evaluate candidate technologies such as by deploying cluster randomized stepped wedge (Hemming et al 2015) or adaptive trial de-signs and techniques (eg response-adjusted randomization) (Pallmann et al 2018) These approaches were also proposed for use in research to assign candidate experimental treatments and vaccines during the 2014-15 Ebola outbreak (Berry et al 2016) When formal research activities are pursued ethics principles and legal requirements for the conduct of research should apply (eg The Belmont Report)
Furthermore any workplace or institution that incentivizes or man-dates use of DCTT has a responsibility to provide evidence that the in-tervention at minimum is not likely to cause harm and to monitor for unanticipated burdens In all cases it is vital that a trusted intermediary be involved in the evaluation of DCTT programs to limit perceptions of bias and ensure a legitimate basis for decision-making Nonsensitive aggregate DCTT analyses should be made available to the public so as to permit verification and inform continuing public debates about its useful-ness and necessity At an individual level data should be available to users that would permit them to further investigate their personal risk with public health officials or other health workers This is important not only to ensure their health and well-being but also to add a layer of protection against unnecessary quarantine
Recommendations
bull Reviews of DCTT systems must be conducted in part by an inde-pendent intermediary that has established the publicrsquos trust
bull Those who authorize use of DCTT within a particular jurisdic-tion or institution should continuously and systematically monitor the technologyrsquos performance in that context This should include monitoring for effectiveness and benefit monitoring for harms and monitoring for the fair distribution of both benefits and harms They should also monitor evidence that is being generated in other contexts about their selected technological solution and about other competing technologies
48 Digital Contact Tracing for Pandemic Response
bull Data should be available to users that would permit them to further investigate their personal risk with public health officials or other health workers to add a layer of protection against unnecessary quarantine
Public Trust and Public Attitudes
Researchers have estimated perhaps conservatively that DCTT use by 80 of smartphone ownersmdash56 of the populationmdashwill be needed to suppress the epidemic (Hinch et al 2020) These estimates also highlight that some decrease in transmission would be realizable even with lower rates of technology adoption As such in order to maximize impact it is essential to gain a thorough understanding of public perspectives on DCTT including which features and uses of the technology the pub-lic finds acceptable which kinds of DCTT the public would be most likely to use and which designs and uses of DCTT would maintain or jeopardize public confidence and trust There will be variation in public attitudes within and across societies and over time
With respect to what we currently know about public attitudes and trust in DCTT in the United States polling data suggest some potential support and also some divisions regarding willingness to use the technol-ogy Polls conducted by groups based at the University of Zurich (Hargit-tai et al 2020) and the University of Oxford (Altmann et al 2020) suggest that more than 60 of Americans would be willing to install such an app Both a Washington PostndashUniversity of Maryland poll (2020) and a Kaiser Family Foundation poll (Kirzinger et al 2020) show roughly half of the population would be willing to install the app Over half of the population (59) would be willing to share their COVID-19 positive test result with an app in order to anonymously share that information with their contacts (Washington PostndashUMD 2020) Only 29 of respondents to a March 12ndash27 Oliver Wyman Forum poll (Elliott et al 2020) said that they would be willing to share their location data Additionally Washing-ton PostndashUMD data and Pew data from 2019 suggest that approximately one in six Americans do not have a smartphone and thus cannot use the technology without intervention (Pew Research Center 2020)
People may be more willing however to download an app if it will
Ethics of Designing and Using DCTT 49
ease social distancing policies and allow for more economic and social activity Willingness to install a contact tracing app increased among re-spondents to the Kaiser Family Foundation poll from 50 to 66 when respondents were asked if they would be willing to do so to allow schools and businesses to reopen Additionally who develops or administers the app appears to matter Respondents to the Washington PostndashUMD poll indicated higher levels of trust that their anonymity would be preserved by public health agencies and universities than by tech companies or health insurance companies Further more respondents to the Oliver Wyman Forum poll were willing to share their health information with public health authorities (55) than the local government (35) their employer or school (33) or the federal government (27)
These data suggest that people will be more willing to use a contact tracing app when the potential benefits are clearly identified and valued such as lifting social distancing measures and they will be more willing to do so if the data are going to a public health agency rather than the federal government or a tech company Other factors that seem to be as-sociated with greater willingness to install a contact tracing app include younger age and the app source (Hargittai et al 2020) with a preference for apps distributed by public health agencies over others such as health insurers or public universities (Hargittai and Redmiles 2020) However all of this must be read with caution as public polling may not be repre-sentative of some populations or of widespread public attitudes Further these attitudes may shift over time and may be discordant with behaviors (Barth and de Jong 2017)
Deliberative public engagement efforts would be an appropriate means of filling in gaps in understanding about the acceptability of dif-ferent approaches (Fishkin and Laslett 2003 Cavalier 2011) In addi-tion including the public particularly in the earlier stages of planning a path to sustainable resolution to the pandemic could serve to help disseminate a nuanced understanding of what is at stake including the key challenges and trade-offs Aggregated public polling results are not sufficient as a proxy for careful analyses of the ethical challenges but they do provide a necessary input for these analyses Integrating lessons and outputs from public engagement into guidance and other products requires special attention and should be validated and enhanced through further engagement
50 Digital Contact Tracing for Pandemic Response
Recommendations
bull More research into public attitudes is needed In particular in-depth qualitative research should examine public attitudes about percep-tions of trust in DCTT among different communities which fea-tures of DCTT influence trust and the extent to which people are willing to provide different types of data through DCTT to help their community
bull States and localities that are considering adopting DCTT should engage with the public to increase their understanding of the ac-ceptability of DCTT design features and uses among diverse communities
Designing Flexible Technology to Maximize Public Health Utility While Respecting Other Values
Values in Design
Efforts to advance DCTT in the United States and elsewhere have empha-sized the importance of ldquoprivacy by designrdquo that is building privacy and security protections into the design of technology rather than counting on responsible use alone (Cavoukian 2010) As noted above some major technology companies have signaled this position through development of decentralized privacy-preserving proximity tracking (PPPT) systems These systems embed features such as decentralization anonymity of us-ers bans on collection of location data and minimal reliance on or inte-gration of public health authorities or other government actors Many of these features have also been embraced early by advocacy organizations (Crocker Opsahl and Cyphers 2020 Electronic Privacy Information Center 2020 Kahn Gilmor 2020) and in an open letter from nearly 300 researchers (ldquoJoint Statement on Contact Tracingrdquo 2020)
Privacy by design provides principles that incorporate one set of val-ues (privacy) into the design of DCTT Importantly the principles ac-knowledge the need to design privacy defaults into systems while main-taining the capacity of those systems to achieve their otherwise justifiable ends Put another way privacy by design ldquoembraces legitimate non-pri-vacy objectives and accommodates them in an innovative positive-sum mannerrdquo (Cavoukian 2010 p 4)
Ethics of Designing and Using DCTT 51
This stance simple in its statement is not easy to satisfy Given that ldquoobjectivesrdquo are themselves driven by values it begs for an articulation of additional values (aside from privacy) that individuals and groups within societymdashincluding many privacy advocatesmdashmay believe to be important For example at any moment in addition to valuing their own privacy individuals may value efficiency equity autonomy economic well-being companionship patriotism or solidarity Moreover the above stance necessitates an acknowledgment that peoplesrsquo value priorities often change when circumstances change not least of which during a pandemic when mass physical distancing has made it difficult to fully realize many important values (aside from physical privacy) A different orientation is needed at this moment As Flanagan Howe and Nissenbaum (2008) conceptualized in 2008 we should take a ldquovalues in designrdquo approach to DCTTmdashan approach that designs a broader range of values such as those enumerated above into technology
This approach requires a wider ethical lens through which to ex-amine DCTT and requires hard but important work to appropriately balance competing interests within technology architecture For example there is value in technology providing users the option to collect their location history and share it with public health professionals in order to advance the public health response increase system efficiencies (eg by contributing information that can lead to better data processing) and reduce the burden on essential workers For some this might be an ex-pression of autonomy solidarity or patriotism At the same time there is value in further advancing autonomy by designing technology to allow individuals to control what data about them are collected and shared
Justifying a Middle-Ground Approach to DCTT
We ought to embrace a DCTT that has a default of interoperability and privacy protection but that does not stop there Triggering events such as entry of a positive test result or receipt of a notification that one was proximate to someone who tested positive could for example generate a push notification that users can acknowledge in order to permit transmis-sion of potentially useful location data to public health authorities This could be accompanied by an explanation of the value of the information and relevant restrictions on its use
At this point it is worth reiterating that manual contact tracingmdash
52 Digital Contact Tracing for Pandemic Response
which involves collecting information from people whorsquove tested posi-tive and their contactsmdashincludes collection of personal information and potentially embarrassing or sensitive data about the places theyrsquove been and the people theyrsquove had contact with Manual contact tracing efforts use these data to uncover ongoing transmission provide useful informa-tion tailored to the individual and enable isolation and quarantine as necessary
It stands to reason that if these forms of data can be collected by a DCTT and provided to public health authorities in a maximally secure and voluntary way (with clear rules regarding authorized uses) this may amplify public health authoritiesrsquo manual contact tracing efforts For ex-ample location data from DCTT could help jog peoplersquos memories about where theyrsquove been and fill in memory gaps This is especially relevant given the long period of infectivity of SARS-CoV-2 which begins before people are symptomatic and therefore before they are aware they are infected (Ferretti et al 2020) Location data might reveal that a COV+ person was at a restaurant at an exact time and date which could be followed up by contact tracers who could alert the public or use other measures to reach those who were also present in the restaurant at the same time In other disease contexts (see Furlanello et al 2002 Dredze et al 2013 Eckhoff and Tatem 2015 Fraccaro et al 2019) geolocation data have demonstrated some potential to support epidemiology and dis-ease surveillance with technical cautions regarding accuracy and the like (Beukenhorst et al 2017)
These benefits are currently speculative for DCTT At present pro-viding public health authorities with large amounts of data on cases and potential case contacts will be useful only if there is sufficient public health capacity to follow up on these data In addition there is a risk of low-quality data from DCTT flooding the system Investigating poten-tial case contacts identified by a DCTT may distract them from other important efforts and at some point overwhelm public health capacity altogether Whether and to what extent data from DCTT will benefit contact tracing efforts is unknown pointing again to the importance of continuously collecting high-quality evidence about DCTT
Nevertheless what would enable the most flexible and potentially robust public health response is to design DCTT so that restricted data sharing is possible From an ethics perspective the collection and use
Ethics of Designing and Using DCTT 53
of sensitive data in manual contact tracing efforts (described above) is typically seen as ethically justifiable so long as there is sufficient public health benefit and need Thus wouldnrsquot it seem appropriate from both a public health and ethics perspective to design DCTT systems to enable similar data to be shared with public health authorities when and if there is ethical justification for sharing them
Why instead do so many advocate that DCTT should be designed as a ldquominimalrdquo system when this arguably ties the hands of public health and individual users and precludes the collection of data that public health authorities (and indeed many other apps on our phones) typically collect We here consider and appraise some of the reasons that may motivate individuals and groups to argue for minimalistic positions
1 Proponents of minimal systems may believe that such systems will be most
widely adopted Some groups have maintained that only these systems will earn and maintain public trust and be widely adopted (Simpson and Conner 2020) For example the previously referenced open letter (ldquoJoint Statement on Contact Tracingrdquo 2020) asserts ldquoSome of the Bluetooth-based proposals respect the individualrsquos right to privacy whilst others would enable (via mission creep) a form of government or private sector surveillance that would catastrophically hamper trust in and acceptance of such an application by society at large It is crucial that citizens trust the applications in order to produce sufficient uptake to make a difference in tackling the crisis It is vital that in coming out of the current crisis we do not create a tool that enables large scale data collection on the population either now or at a later time Thus solutions which allow reconstructing invasive information about the population should be rejected without further discussionrdquo
Response While it is true that public trust in and acceptance of DCTT is essential for its success there is insufficient evidence that public trust would be threatened by a DCTT system that has the capacity to collect location data and enable voluntary sharing of those data with public health authorities A contrasting perspec-tive is that maintaining public trust requires maintaining public confidence that the DCTT system is providing useful information is benefiting and not harming individuals and is advancing the
54 Digital Contact Tracing for Pandemic Response
public health response (Leprince-Ringuet 2020) From this per-spective a system that is less well integrated into the broader public health response or that generates a higher rate of false positives (as some suggest decentralized approaches might (Fraser et al 2020)) may fare worse when it comes to maintaining public confidence and trust
2 Proponents may hold the view that minimal systems are harmless (or nearly
harmless) to individuals This is because individuals are anonymous none of their location data are gathered and none of their identifiable data are shared with anyone In contrast DCTT systems that collect and share identifiable data including location data may be seen as posing risks of harm to individuals
Response While minimal systems may be harmless (or nearly harm-less) from the perspective of protecting privacy they may not be harmless from the perspective of public health if they generate system inefficiencies through producing too many false positive or false negative contacts Aside from presenting a challenge for public health professionals false positives could also harm individ-uals If users receive a large volume of automated messages alert-ing them to proximity events will this cause distress Will a large volume of alerts cause users to become disengaged and stop using the DCTT or lose confidence in contact tracing more generally as a legitimate method of disease control Admittedly these are just potential harms and risks it is unknown the degree to which they will materialize The point is that privacy-related harms are not the only relevant harms to individuals that we should consider when assessing DCTT
We acknowledge the risk under a middle-ground DCTT of data being used in ethically unjustifiable and harmful ways For exam-ple it would be against the principles and recommendations artic-ulated in this report for data to be sold or monetized by technology companies or others for corporate gain and this misuse of data would be more intrusive if the data were potentially identifiable What makes it ethically justifiable to take this risk is the compen-sating benefit of allowing the most flexible and robust public health
Ethics of Designing and Using DCTT 55
response during the pandemic but this alone is not sufficient The risk of inappropriate uses must be reduced by ensuring stringent requirements for data security and access as well as clear legal protections and recourse for any violations (as discussed further below)
3 Proponents may believe that DCTT systems should not collect location data
as this would be too intrusive and of insufficient value Some proponents of PPPT systems maintain that recording proximity events is sufficient and data relating to usersrsquo movement and location should not be collected (Ingram 2020) The thought may be all we need to know is whether two individuals came into close enough contact for viral transmission to have occurred we donrsquot need to know where or when this contact occurred and there is no need to collect and store usersrsquo location data
Response This conclusion might be too hasty As discussed above there is potential (though unproven) benefit to providing public health authorities with location data Location data could help jog peoplersquos memories about where theyrsquove been provide more context for understanding the nature of ldquoproximity eventsrdquo captured by the DCTT and allow public health authorities to quickly define a category of individuals who may be at risk Collecting location data from cases is what public health authorities do on a regular basis following best practices for manual contact tracing
In addition many peoplersquos location data are currently gathered by apps on their phones and used for various purposes such as to provide more accurate navigation to offer entertainment or to improve services Many are willing to accept these capabilities because they provide some value in return Why not allow DCTT to also collect these data so that the data are available for users to share with public health officials who can then do their work more effectively and refine their understanding of how the disease transmits If many are willing to have these data used to find a bet-ter route home why not let individuals share these data to support the effort to save lives
56 Digital Contact Tracing for Pandemic Response
4 Proponents may hold the view that minimal systems pose little or no threat
to individual autonomy whereas systems that collect identifiable data and
integrate public health do pose a threat to individual autonomy For exam-ple they may worry that use of DCTT could be mandated and not a voluntary choice and in this circumstance mandatory use of minimal DCTT would be less intrusive risky and privacy violating Another worry might be that itrsquos theoretically possible that DCTT could share individualsrsquo data with public health authorities without usersrsquo full understanding if the technology does not even gather identifiable data then itrsquos not possible for these data to be shared without the individualrsquos consent
Response We discuss the importance of appropriately designed disclosures and consent below as well as the high bar that would need to be met to ethically justify mandatory use At this time mandated use of DCTT by states or institutions is not justifiable given uncertainty about potential harms and benefits Users should have a meaningful opportunity to review and understand infor-mation about the specific technology and its uses and to consent Assuming that individuals are not required to use DCTT and that they provide consent to using it designing DCTT to make data collection and sharing possible is the design choice that maximizes individual autonomy because it provides individuals with options they may value
Individuals may wish to share their data with public health au-thorities for both self-interested and altruistic reasons For exam-ple someone who has tested positive for SARS-CoV-2 and enters this test result into an app may wish to be connected to public health authorities in order to be provided with needed information resources and support She may wish for public health authorities to be provided with her phone number in case they need to reach her to provide additional information Further someone who has been alerted by an app that he had a ldquoproximity eventrdquo with a person who has tested positive for SARS-CoV-2 may wish he had location data to share with public health authorities in order to help ascertain whether this event is a cause for concern or whether it is likely a false positive (eg he and the COV+ person were sepa-
Ethics of Designing and Using DCTT 57
rated by a wall) Someone who tests positive for the virus may also wish to share their location history with public health authorities in order to be as helpful as possible to the overall public health re-sponse by facilitating de-identified aggregate analyses that identify locations of higher transmission or contribute to refining overall understanding of the disease and pandemic
5 Concerns about ldquosurveillance creeprdquo and the long-term downstream effects of
digital contact tracing system may also motivate embrace of minimal DCTT Digital contact tracing technology that collects identifiers and loca-tion data and has the capacity to share them with public health au-thorities may represent a massive and concerning increase in govern-ment surveillance of the public It might be feared that the use of this surveillance capacity in the COVID-19 response sets an unwelcome precedent for future use in other contexts Designing DCTT as min-imal systems may be a way to minimize the risk of surveillance creep and to minimize the harms associated with potential future uses of the technology
Response Surveillance creep is a serious concern To guard against surveillance creep protections should be put in place to ensure that only those data that are necessary and relevant for the public health purposes at hand are collected and used and data should be kept only for the period of time needed for those public health purposes In the face of these concerns it is important to emphasize that widespread use of DCTT in the COVID-19 response is justi-fied by the exceptional circumstances of the current pandemic and their use in this context does not imply that future public health use is ethically appropriate without significant public debate (eg use in seasonal flu surveillance efforts) Future use will require in-dependent justification Use of DCTT in other contexts (eg law enforcement or immigration enforcement) is also presumptively unethical
All in all the arguments that DCTT should be designed as a minimal system are not convincing Rather DCTT should be developed through a ldquovalues in designrdquo approach with a core set of features that protect pri-
58 Digital Contact Tracing for Pandemic Response
vacy with enough flexibility to be used differently depending upon local conditions evolving evidence and individual preferences What kind of digital contact tracing system will strike the right balance between public health goals and other considerations will depend upon circumstances For example whether it is even beneficial to provide public health au-thorities with volumes of data about potential contacts of COV+ people will depend in part upon whether they have the capacity to make good use of those data This will vary from location to location and will change over time
Recommendations
bull Technology companies should not alone control the terms condi-tions or capabilities of DCTT nor should they presume to know what may be acceptable to members of the public
bull A ldquovalues in designrdquo approach to development of DCTT should be adopted (Flanagan Howe and Nissenbaum 2008 Knobel and Bowker 2011) Robust public- and user-engagement activities should be pursued to identify and incorporate to the extent pos-sible a range of values into the design of the technology These values may include privacy but also autonomy efficiency equity or others Technology design should reflect an appropriate balance and prioritization of identified values
bull Technology design should not be static but rather it should be ca-pable of evolving depending upon local conditions new evidence and changing preferences and priorities
bull DCTT should be designed to have a base set of features that protect privacy with layers of additional capabilities that users may choose to activate An initial default should be that user location data are not shared but users should be provided with easy mechanisms and prompts to allow for opting-in to this capability with encourage-ment to the public if and as it is shown to be critical to achieving public health goals
Ethics of Designing and Using DCTT 59
Policy Positions to Advance Widespread Use of Digital Contact Tracing Technologies
The public health value of a DCTT depends in part on the number of people who use it This section concerns broad public policy positions that relate to the widespread adoption of DCTT What are ethical means of encouraging or securing widespread adoption of DCTT systems Un-der what circumstances would it be ethical to mandate their use or incen-tivize their use What enforcement challenges exist
Mandating Use
Digital contact tracing has occurred without the publicrsquos explicit volun-tary agreement in some countries such as China and Israel In others use has been voluntary (Valentino-DeVries Singer and Krolik 2020) For example Singapore adopted an app that the public could use on a vol-untary basis and approximately 20 of the population has downloaded and used it Norway has recently launched a contact tracing app that was downloaded by roughly 30 of the population in the first week that it was made available In the United States many advocates and researchers have argued that use of digital contact tracing tools must be fully volun-tary this is the dominant perspective
There are numerous ways that DCTT could be put into use without user choice For example as has been done in Israel location data from mobile phones could be collected and used by the government without usersrsquo consent Use of an app could be formally mandated as a precon-dition for returning to work or school or even further to control entry into a facility or onto transportation such as airplanes through scanning of a QR code to demonstrate personal exposure levels (Gan and Culver 2020)
While these approaches are hard to imagine in the United States some contend that mandatory use of digital contact tracing tools could be ethical and may even be ethically required Mandating use of digital contact tracing tools could in theory vastly increase the effectiveness of digital contact tracing systems and thus may save more lives and allow states to lift lockdowns sooner or avoid reimposing lockdowns in the future Canca (2020) argues that use of privacy-by-design digital contact
60 Digital Contact Tracing for Pandemic Response
tracing tools should be mandatory because the use of these tools will be nearly harmless if there are sufficient privacy protections In addition mandatory use of DCTT that embraces these principles is significantly less intrusive at the individual level than manual contact tracing which involves the collection of personally identifying and potentially sensitive data In this light it could be argued that such mandates are actually pref-erable from the perspective of both public health and individual liberty insofar as they reduce the likelihood of ldquostay at homerdquo orders which are a severe limitation of individual liberty
Nevertheless mandated use of DCTT systems faces considerable ob-stacles For example people may not adhere to the mandate by simply leaving their phone at home thus preventing their activities from being tracked Even more harmful would be if people react to a mandate and a perceived violation of liberty and privacy by employing location and Bluetooth spoofing software to shield their real contacts behind a screen of misinformation The introduction of this misinformation into a contact tracing effort might severely undermine its effectiveness The possibility of nonadherence also raises the issue of enforcement would high rates of nonadherence be permitted or would enforcement be attempted (if even possible) Perhaps more important should the technology not deliver the hoped-for benefits having mandated the use of an unproven technology could result in a loss of public trust in the technology the entity instituting the mandate and potentially the larger public health response (Bernstein et al 2019)
Mandatory DCTT could also be used to enforce quarantine restric-tions and stay-at-home orders for those who are COV+ or are determined to be at heightened risk The use of DCTT in enforcement activities raises a number of ethical (and legal) issues that are beyond the scope of the present analysis In particular individuals have a heightened interest in personal privacy if their data can be used to restrict their freedom of movement and other civil liberties At a minimum stringent procedural protections would be required to ensure that the data collection is fair and unbiased and that DCTT users are provided with adequate informa-tion in advance about how their data may be used
Mandatory use policies for DCTT must therefore convincingly ad-dress a number of questions including
Ethics of Designing and Using DCTT 61
bull Is the technology designed to meet an important and unmet public health need
bull Is there sufficient evidence to suggest that the technology will be effective at serving its purpose
bull Is the outbreak characterized by sufficiently severe morbidity and mortality and a high rate of disease transmission
bull Are there other less autonomy-restricting or less risky alternatives to widespread mandatory use of DCTT
bull Is it possible and likely that a sufficient number of individuals will comply with a mandate
bull Can inequities in the burdens and benefits of the mandate be suffi-ciently addressed through social protections and countermeasures
bull Can enforcement and enforcement discretion be implemented in a manner that is consonant with fundamental rights
bull Will those subject to the mandate interact closely with a population that is at high risk of morbidity or mortality if they contract the virus
bull Is it possible to mandate use and remain consistent with important ethical and legal principles
These questions would need to be satisfactorily addressed and explicitly documented by any decision maker considering mandatory use includ-ing government officials institutional leaders and employers Particularly important is the need to identify reliable evidence that the DCTT would be effective and to ensure that the burdens and benefits of use are equita-ble and justifiable If use of a DCTT is a condition for returning to work or school in person those who refuse or are unable to use DCTT should not lose their jobs or positions as a result and adequate support should be in place for people who are asked to self-quarantine
Finally it is important to distinguish a mandate from a ldquopushedrdquo program installation or a default setting in an application which can be modified by users A mandate relates to a policy of required use whereas the pushed programs or default settings relate to the chosen architecture for download and operation of the application
62 Digital Contact Tracing for Pandemic Response
Incentivizing Use
Perhaps the most effective way to generate widespread adoption of DCTT in the United States is to offer incentives to individuals who choose to adopt and who properly utilize the preferred DCTT approach in a voluntary system External incentives may help ldquonudgerdquo populations toward desired adoption targets Given the importance of widespread use of DCTT modest incentives ought to be considered for DCTT in the US if and when there is sufficient evidence of the technologyrsquos utility Note that in other contexts studies have shown that the provision of some incentive leads to an increase in adoption or utilization of public health programs (Singer and Ye 2013 Lee et al 2014) Moreover even a relatively small incentive can achieve much greater rates of adoption with some studies demonstrating that the incremental adoption gain de-creases as the incentive gets larger (Thornton 2008 Gibson et al 2019) In the context of COVID-19 incentives that might be both effective and ethically acceptable could include a relatively small monetary token free or discounted mobile phone service for a period of time or credit to be used by means of a mobile phone
Not all incentives are ethically appropriate For example making access to lifesaving health care contingent on using a DCTT or making valuable disease information available only to DCTT users would not be ethically appropriate In addition incentives cannot be used to over-come otherwise ethically unjustifiable technology design for example they should not be used as an offset for providing personally identifiable health information to other users
Importantly incentivization schemes must be kept distinct from man-dates as the latter require greater ethical justification To offer an incen-tive is to offer something of actual value to individual participants over and above what they are reasonably entitled to at baseline For example making a return to work contingent on using DCTT is not offering an incentive but instead imposing a mandate and it would have to be justi-fied as a mandate
In the context of COVID-19 it is also necessary to recognize that there is an inherent ldquoincentiverdquo behind the technologymdashthat is the prom-ise of more lives saved faster pandemic recovery and the reduction or elimination of blanket physical distancing Effective public communica-
Ethics of Designing and Using DCTT 63
tion of these goals if and when there is sufficient confidence in the tech-nology is important
Encouraging Use
Another important approach to increasing use of DCTT in the United States is for trusted leaders to encourage their use Community leaders public figures health care professionals and other respected individuals who have the publicrsquos trust and goodwill could be enlisted to commu-nicate with the public about DCTT and encourage its use drawing on notions such as communal responsibility solidarity and so on These en-couragements could be combined with other approaches (eg small in-centives) to optimize reach while continuing to respect individual choice
Recommendations
bull DCTT use should not be mandated at this time given uncertainty about potential harms and benefits Additional technology user and real-world testing is needed
bull Incentives can be a useful complement to encouragements how-ever any incentives for users to install and use DCTT must be equitable should not be coercive and should align with effective use of the technology (ie they should not incentivize downloading an app but then leaving onersquos phone at home)
bull Trusted leaders should be enlisted to communicate effectively with the public about DCTT and encourage its use should the technol-ogy demonstrate some potential The limits of knowledge regard-ing effectiveness should also be explained along with what will be done to improve technological capabilities as understanding evolves
Disclosure and AuthorizationConsent
In deciding whether to use DCTT voluntarily individuals must be suf-ficiently informed both through broad coordinated public engagement campaigns and individual-level disclosures and there must be a meaning-ful mechanism for users to consent It is important to recognize that while
64 Digital Contact Tracing for Pandemic Response
informed consentmdashwhich is characterized by detailed consent forms and requires a witnessed signaturemdashis the standard for most research and clinical care encounters (Faden and Beauchamp 1986) it is not typically the standard for public health disease surveillance In the public health context other relevant protections (such ethics training for public health professionals and strict data handling and confidentiality requirements) are in place and there is a strong public health interest in collecting the relevant data A more limited role for consent has been recommended for public health surveillance based on a reciprocal obligation of members of society to contribute to a ldquocommon goodrdquo and particularly in the con-text of a pandemic practical considerations such as time constraints and exigencies such as increasing morbidity and mortality (WHO 2017)
Under current circumstances given that (1) many individuals have time and capacity to consent (2) DCTT is being considered as part of plans for longer-term restabilization (3) DCTT is not a familiar part of our public lexicon (4) remote consent disclosure and authorization can be easily embedded in DCTT systems (Moore et al 2017) and (5) there are justifiable public deficits in trust with respect to various government and corporate actors handling potentially personal digital information a strong ethical case can be made for requiring a carefully crafted version of what is sometimes referred to as simple consent Simple consent consists of basic disclosure and voluntary agreement or authorization (Ali et al 2017) Three questions then arise
1 What information should be disclosed to potential users of DCTT
bull Information disclosed might include
deg Entity responsible for the technology
deg Its purpose
deg How it works (in lay terms)
Some participatory disease surveillance systems (eg Flu Near You) have received for-
mal ldquowaiversrdquo of consent requirements from institutional review boards (IRBs) in the US
As they undergo development these digital surveillance systems often straddle a line be-
tween public health surveillance and research hence the frequent need or desire to obtain
ethical review by an IRB (Ali et al 2019)
Ethics of Designing and Using DCTT 65
deg What users need to do
deg Any user options eg
Sharing geolocation data with public health authorities when that would facilitate a defined public health goal
Sharing de-identified metadata with technology develop-ers (for system enhancement)
deg User rights
deg How data will be handled
What data are collected
What data are shared (and how and with whom)
Purposes for which data can be used and not used
How data are secured and protected
Whether and what data will be retained (or will be deletable)
deg Potential benefits and any known risks
deg How to obtain answers to questions about the technology and public health response
2 How should this information be presented
Information should be presented leveraging eConsent models that are more accessible than long ldquoclickwraprdquo disclosures typical of mobile apps (Iwaya et al 2019) For example a simple open-source smartphone con-sent module that has been developed by Sage Bionetworks for research uses could be adapted to the public health surveillance context and to DCTT (Doerr Suver and Wilbanks 2016)
bull Formatting recommendations include (cf Doerr et al 2016)
deg simple and straightforward information
deg deliberately organized content
deg multimodal learning (eg visual audio written)
deg accessibility for disabled users
66 Digital Contact Tracing for Pandemic Response
deg multilingual text
deg engagement through interaction (eg swiping to navigate forward and backward)
bull The same simple information should be made publicly available via multiple other platforms (eg on websites in newspapers over social media)
bull More detailed disclosures should be made readily accessible to those who wish to learn more with no hidden surprises
3 How should users signal that they agree to the details specified in disclosures
Opt-in Models
Opt-in models are those that through an affirmative act such as clicking a button users would indicate their intention to use a DCTT This ap-proach is consistent with other app downloads where app details and privacy policies are made available through a download page and users are required to affirmatively click a button to install an app Once in-stalled some apps further alert users to particular ways in which phone capabilities or data will be used with some permitting selective toggling (opting-in or opting-out) of certain features With DCTT apps in addi-tion to disclosures provided on a download page the user could be guided through a simple interactive module embedded in the app (such as is described above) in order to increase the chance of meaningful exposure to important information about the technology and how data will be handled At that point any user options such as those itemized above could be described and choices made
Opt-out Models
There are at least two different ways in which the term ldquoopt-outrdquo has been used in this context The conventional use of the term ldquoopt-outrdquo is characterized by an act which signals an individualrsquos intention to decline something that would have otherwise occurred without intervention A few others have used the term to refer to ldquorevocation of consentrdquo for example the United States COVID-19 Consumer Data Protection Act of 2020 Senate bill (S3663) would establish a default opt-in positionmdash
Ethics of Designing and Using DCTT 67
requiring ldquoaffirmative express consentrdquo for collection and use of prox-imity and other related datamdashand refers to individuals having a right to later revoke their consent through an ldquoopt-outrdquo The latter use of the term is not our focus here
Given this a DCTT app that is voluntarily downloaded through an affirmative act would be difficult to characterize as an opt-out approach This leaves more passive surveillance systems that rely on automatic in-stallation of self-activating technology onto users phones There are a range of views among the authors of this report about the value of an opt-out approach for DCTT with some arguing for an opt-out approach on grounds that it might increase coverage and would be ethically acceptable if accompanied by similar disclosures as above to ensure users are aware of the technology and data uses (Mello and Wang 2020) This approach would present users with a mechanism to opt-out if they wish which should be reasonably easy to effectuate Under these circumstances as noted above an ldquoopt-outrdquo would not be synonymous with mandating use of the technology
Others among the authors argue that there is reason to believe that opt-in approaches may be able to sufficiently achieve desirable levels of utilization relative to opt-out approaches Unfortunately data related to opt-in versus opt-out models of DCTT are very limited One recent sur-vey (Altmann et al 2020) found that across five countries (UK Germany France Italy US) slightly more people reportedly would download an app under an opt-in system (748) than would keep an app on their phone under an opt-out system (677) Moreover when US respondents were directly asked which approach they would prefer 60 indicated a preference for opt-in This remained true across various demographic variablesmdashgender region political affiliation lockdown status and other characteristics Whether actual behaviors would align with anticipated behaviors in the context of DCTT remains an unanswered question that should be carefully studied under real-world conditions There are a range of important empirical questions regarding how much and what kind of impact (positive or negative) various types of defaults might generate for public health and for different mobile phone user groups including vul-nerable and marginalized users
Opt-out models for app authorization may encounter greater legal
68 Digital Contact Tracing for Pandemic Response
and political challenges especially if the COVID-19 Consumer Data Pro-tection Act of 2020 (S3663) the competing Public Health Emergency Pri-vacy Act (S3749) or another similar bill is enacted in the United States Both of these standing bills require affirmative opt-in consent Opt-out approaches also risk negative reactions from some mobile phone users a small number of whom may go so far as to intentionally interfere with data because of the perceived intrusiveness of an automatically installed tracking platform (Dixit 2020)
Given these considerations and the apparent willingness of a large portion of the population to opt-in to use DCTT an opt-in approach to authorization should be instituted to accompany initial DCTT rollout The feasibility and value of opt-out approaches should continue to be evaluated informed by what is technologically possible what local as-sessments of benefits and harms of the technology reveal over time and our evolving understanding of the degree to which an opt-out approach is likely to increase or decrease utilization Opt-out approaches should not be precluded
Recommendations
bull A clear and concise module consisting of basic disclosure and vol-untary authorization should be developed to accompany DCTT This module should not take the form of ldquoclickwraprdquo terms of service or end-user agreements but rather provide only essential information necessary for an individual to make a decision More detailed disclosures (such as FAQs in plain language) should be made easily accessible to those who wish to learn more with no hidden surprises
bull An opt-in approach to authorization should be instituted to ac-company initial DCTT rollout The feasibility and value of opt-out approaches should continue to be evaluated informed by what is technologically possible what local assessments of benefits and harms of the technology reveal over time and our evolving under-standing of the degree to which an opt-out approach is likely to increase or decrease utilization among different populations Opt-out approaches should not be precluded
Ethics of Designing and Using DCTT 69
Promoting Equity and Fairness
Digital contact tracing technologies should be designed and used in ways that as far as possible promote an equitable distribution of benefits and burdens DCTT should be deployed in a manner that does not propa-gate preexisting patterns of unfair disadvantage or distribute harms and risks unfairly throughout the population For example communities with lower rates of technology and data access may benefit less from DCTT Special attention must be paid to communities that experience preexisting health disparities and to those that are being hardest hit by the pandemic
Digital Disparities
In the United States February 2019 data indicate that approximately 80 of the population are smartphone users (Pew Research Center 2020) though rates of mobile phone use are significantly lower among people over age 65 (53) people with any disability (58 2016 data) (Anderson and Perrin 2017) people with less than a high school edu-cation (66) people who earn less than $30000 per year (71) and people who live in rural areas (71) As a result these populations and communities may use DCTT in lower numbers thereby lessening the effectiveness of DCTT and the likelihood of benefit for these populations from such systems Moreover it has been reported that many older and less costly smartphones (roughly estimated at 10ndash20 of smartphones in the US) lack important capabilities required for the leading AppleGoogle platform to work (Bradshaw 2020) This is of special concern because some of the above groups that are less likely to own smartphones in general are also less likely to own newer smartphones with the needed capabilities Some within the above groups (eg people who are older and people identified as Hispanic African American or American Indian) are also disproportionately experiencing morbidity and mortality from COVID-19 (CDC 2020h)
One may argue that by using DCTT human and financial resources that would otherwise be spent on manual contact tracing will be pre-served and these resources can then be redirected to better meet the needs of those who are not otherwise being effectively served by the technology because of disparities or for other reasons This argument has intuitive
70 Digital Contact Tracing for Pandemic Response
appeal and should be taken seriously however it is unsettled whether DCTT will contribute sufficient efficiencies to the overall public health response to make it possible financially and logistically for manual ser-vices to be allocated in greater proportion to those who are unable to benefit from DCTT It is entirely possible that at least in the short-term DCTT may introduce new inefficiencies due to unintended consequences or the need for public health officials to follow up many more contacts One possible mitigation to the challenge of digital disparitymdashthough it does not solve the underlying challenge of ensuring net efficiency across systemsmdashmight be to provide mobile phones or other devices and data packages to those who would otherwise be left out
Disparate Risk of Harm from Surveillance and Data Gathering
Ensuring wide digital coverage does not however resolve other equity concerns It is important to consider that some populations may experi-ence greater harm and fear of harm from having their data collected For example some groups such as African Americans Hispanic Americans Muslim Americans and undocumented immigrants have more reasonable fear of their data being handed over to law or immigration enforcement and some groups have lower levels of trust in public health due to past injustices (CSM 2017 Pew Research Center 2017 Rodrigues et al 2018 Auxier et al 2019) Any data gathered by DCTT should be used solely for public health purposes Efforts should be made to assure members of these and other communities that their data will not be misused or made available to those outside of a public health context In addition if DCTT are used in the current pandemic this should be with the understanding that future use of DCTT in other contexts (eg law enforcement or im-migration enforcement) is presumptively unethical
Some preliminary polling related specifically to DCTT emphasizes the complexity of the challenges faced and the need for deeper public engagement (Anderson and Auxier 2020) The polling results suggest that people who identify as African American or Hispanic are more likely than people who identify as White to consider government tracking of mobile phones as acceptable These findings like many others are difficult to in-terpret given background political polarization on the issue More direct engagement is required to better understand how different communities comprehend and experience DCTT and other forms of surveillance
Ethics of Designing and Using DCTT 71
Discrimination and Stigma
Stigma may result from an individual being identified as COV+ or a neighborhood or establishment becoming identified as a ldquohotspotrdquo as a result of numerous COV+ people living in that area or having visited that establishment In particular certain groups may suffer more as a result of being associated with COVID-19 such as the well-documented blame that has been directed toward Chinese people (and broadly East Asian communities) or the communities that are disproportionately likely to contract the illness (Devakumar et al 2020) When identifiable lo-cation data are made public as has been the case in South Korea per-sonal and private information were revealed Furthermore businesses in South Korea that were identified as having patrons who tested positive for COVID-19 have suffered economic losses and stigma (N Kim 2020)
To avoid the stigma and potential discrimination that can result from being identified as COV+ DCTT must never make data publicly avail-able that could be used to identify persons who have tested positive Safe-guards must be in place to ensure that any identifiable data that may be gathered for public health purposes are protected If DCTT data are used to provide heat maps to the public of locations that COV+ individuals frequently visit so as to provide representations of geographic risk or for other reasons it is essential that care be taken to avoid unfairly distrib-uting further economic burdens or other stigmatizing and discriminatory outcomes
Recommendations
bull A commitment to equity means a commitment to ensuring that the benefits and burdens of DCTT are distributed fairly Public engagement is an important tool for assessing impact and to rectify inequities
bull States localities and institutions that recommend widespread use of DCTT should provide technology (eg mobile phones Blue-tooth devices) and free data packages to those who desire but lack access to these devices
bull If there are lower rates of adoption of DCTT systems in some identifiable communities public health authorities should iden-
72 Digital Contact Tracing for Pandemic Response
tify ways to compensate For example directing more non-DCTT resources and efforts toward those communities to meet specific needs that are elsewhere being supported by technology
bull If maps are generated based on DCTT to provide the public with the locations that COV+ individuals have visited steps must be taken to minimize the stigma and potential financial losses that could result from being identified as a hotspot
Instituting Transparent Governance and Oversight
DCTT must be developed with an eye toward both present and future implications Transparent and publicly trustworthy management gover-nance and oversight of DCTT technology and data is both a near- and long-term necessity We face significant uncertainties DCTT technologies are rapidly developing Their risks capabilities effectiveness and down-stream implications are not yet well understood
Concerns about ldquoSurveillance Creeprdquo
Significant concerns have been expressed by privacy advocates (Guari-glia 2020) and in the popular press (Giglio 2020) about what is known as ldquosurveillance creeprdquo Their worry is that state and corporate actors will use new surveillance technologies capacities and permissions well beyond the purposes for which they were initially justified to the public and beyond the time when they are useful for the COVID-19 pandemic
Surveillance creep should be guarded against Only those data that are necessary and relevant for the public health purposes at hand should be collected and used and data should be kept only for the period of time needed for those public health purposes Data should be used only for public health purposes
Any use of DCTT during the current pandemic would be justified by the circumstances of this pandemic and its use in this context does not set a precedent for future public health use (eg use in seasonal flu surveillance efforts) Future use will require independent justification Use of DCTT in the future in other contexts (eg law enforcement or immi-gration enforcement) is presumptively unethical
Broadly speaking efforts should be made to generate public aware-
Ethics of Designing and Using DCTT 73
ness and consensus that use of DCTT in COVID-19 efforts does not imply that future use is justifiable However generating this public aware-ness may be particularly challenging given the complexity of the informa-tional environment where public debate ranges from legitimate concerns about surveillance creep to conspiracy theories regarding the origins of the COVID-19 pandemic (Muller 2020) This means authorities bear spe-cial obligations to be clear on how they plan to use the technologies what oversight mechanisms will be employed to address potential abuse and how they intend to publicize the conditions under which programs will be terminated making sure they are followed
Oversight and Ethical Review
We are rapidly gaining knowledge about SARS-CoV-2 and COVID-19 but we still have essential gaps in our understanding In the United States public health responses including DCTT will generally be developed and coordinated by individual states regional consortia and associations (Reston Sgueglia and Mossburg 2020) Good governance in this con-text requires transparency and the creation of oversight bodies with the appropriate expertise and representation to allow nimble and effective responses while serving as trusted representatives
To address the range of ethics-related concerns about the design and use of DCTT digital surveillance oversight committees should be estab-lished perhaps at a state level and with a platform for national coordina-tion These committees can provide ethical and regulatory review prior to and concurrent with widespread use of DCTT These committees should be composed of a diverse group of experts capable of evaluating a DCTT system locally including members of communities that experience higher rates of digital disparity
When assessing the design and use of digital contact tracing systems these committees (and the public more widely) should consider not only the risks and benefits accrued during the COVID-19 pandemic but also implications for the future What kind of precedent might use of these technologies during the current pandemic set for future use capabilities in other infectious disease outbreaks or in other social contexts (eg law enforcement) How can we navigate safe use of these technologies in a way that preserves public trust in them and enables the possibility of future beneficial use
74 Digital Contact Tracing for Pandemic Response
As a start it should be emphasized that principles offered in this and other guidance documents do not apply only during the pandemic Future efforts to advance DCTT capabilities during quieter times should make every effort to follow them
Recommendations
bull Digital surveillance oversight committees should be established expeditiously with diverse and qualified membership to provide ethical and regulatory review prior to and concurrent with wide-spread use of a DCTT system
bull Understandable and publicly accessible rules must guide the col-lection access control use storage and combination of data by government authorities public and private institutions and other parties such as public health researchers
bull Only those data that are necessary and relevant for the public health response to COVID-19 should be collected and used
bull Identifiable data should be kept only for the period of time needed for the public health response to COVID-19
bull Identifiable data collected as part of this response should not be shared with anyone other than the relevant public health authorities without additional specific informed consent of individual users
bull Before a government or institution adopts a digital contact tracing program they should state the conditions under which the digital contact tracing program will be terminated
bull Future use of DCTT to advance public health or other efforts (eg use in seasonal flu surveillance) would require independent justification DCTT designed for public health use should not be used by law or immigration enforcement
bull The principles offered in this guidance document apply both during and following the COVID-19 pandemic
75
The implementation of digital contact tracing technology (DCTT) is likely to implicate a number of US laws at both the federal and state levels This section focuses primarily on federal laws as these laws apply nationwide and generally preempt conflicting state laws A comprehen-sive assessment of the legality of any particular DCTT program would require case-specific analysis and attention to relevant state laws includ-ing any that specifically address DCTT which may soon exist in one or more states The analysis here is limited to the United States foreign and international laws will not be addressed
Many of the laws discussed in this section are privacy laws designed to protect individuals from the harms that may result from the unautho-rized or improper use of their personally identifiable information (PII) Under these laws legal concerns will generally be minimized if privacy protections are built directly into the DCTT technology (eg ldquoprivacy by designrdquo) As a general principle DCTT should be designed to collect and store only as much PII as is necessary to achieve the public health purpose Collecting only proximity data for example is likely to raise fewer legal concerns than collecting both proximity data and geolocation data Likewise creating aggregated anonymized or de-identified data will raise fewer legal concerns than using and disclosing PII
As we have argued elsewhere in this guidance document however the public health and societal crisis caused by COVID-19 may justify
Legal Considerations
FOUR
76 Digital Contact Tracing for Pandemic Response
greater encroachments on individual privacy than would otherwise be permissible Regardless of the type of data collected privacy concerns will be reduced if users are afforded the right to choose whether their PII is collected and how it is used and disclosed As such DCTT should gener-ally secure meaningful user consent before collecting PII a process which typically requires both disclosure of relevant information and agreement on the part of the user
Privacy concerns will also be reduced if the use of PII is strictly lim-ited to tracking and limiting the spread of SARS-CoV-2 The use of DCTT data for other purposesmdashsuch as commercial or law enforcement pur-posesmdashwould raise additional legal and ethical concerns In addition DCTT developers may be required to implement governance policies that ensure the secure storage of PII limit data retention periods require transparency about data sharing and maintain records of responses to data requests from government authorities
In short the legality of a DCTT program under current United States law will depend on a number of factors including what type of data is collected how the data are used and who may access them how user consent is obtained whether the entity collecting and using the data is the government or a private corporation the context in which data are collected (eg employment education or commercial) and which states have jurisdiction over the program
Privacy law in the United States unlike in other jurisdictions such as the European Union (EU) and Australia is generally sector-specific and limited in scope The result is a patchwork of protections that differ significantly depending on the entity that collects the data and the type of data collected For example under current law telecommunication carriers are governed by different privacy rules than mobile broadband providers Given the complexity of existing federal privacy law we be-lieve that it would be beneficial for the US Congress to enact new privacy legislation that is specifically tailored to the use of DCTT in response to COVID-19 Congress appears poised at least to debate such legislation a pair of bills recently introduced in the Senate and one in the House of Representatives would significantly restrict the collection of PII by digital devices for COVID-tracing purposes S3663 S3749 HR 6866 116th Cong (2020)
Legal Considerations 77
Data Privacy and Data Security Laws
Telecommunications
A DCTT provider that collects data from a userrsquos mobile phone may be subject to the privacy rules governing telecommunication carriers which are enforced by the Federal Communications Commission (FCC) The data protected under these rules are limited however to certain types of PII termed ldquocustomer proprietary network informationrdquo (CPNI) More-over the rules generally apply only to telecommunications carriers and interconnected VoIP (Voice over Internet Protocol) providers
In particular under section 222 of the Communications Act of 1934 47 USC sect 222 and the implementing regulations of the Federal Com-munications Commission (FCC) telecommunications carriers and VoIP providers must establish and maintain systems designed to ensure that they adequately protect their subscribersrsquo CPNI and they are generally restricted from using or disclosing CPNI without the customerrsquos consent (unless the use of disclosure is needed to provide the services subscribed to by the customer) If customer consent is sought to use or disclose CPNI individual notice must be provided to the customer and such notice must provide sufficient information to enable the customer to make an in-formed decision as to whether to permit the requested use or disclosure
CPNI is individually identifiable information that carriers and pro-viders have collected about their customers including phone numbers called and the frequency duration and timing of such calls Of most relevance to DCTT a recent FCC Notice of Apparent Liability asserted that user geolocation data collected by mobile phone network carriers qualify as CPNI under sect 222 and related rules 35 FCC Rcd 1785 (2) (2020) Pursuant to this notice the FCC fined T-Mobile for selling to third parties location data that were derived from the communication between the mobile phones of T-Mobilersquos customers and nearby network signal towers (The FCC also levied fines against ATampT Verizon and Sprint on the same grounds (Valentino-DeVries 2020)) While the FCC has made its position clear that geolocation data are CPNI courts have yet to weigh in on the matter
Even if geolocation data are CPNI however the FCC can enforce sect 222 of the Communications Act only against telecom carriers and VoIP
78 Digital Contact Tracing for Pandemic Response
providers not against cable broadband and mobile broadband internet providers 47 USC sect 53(44) 47 CFR sect 93 In 2018 the FCC promul-gated a regulation stating that contrary to its prior position its sect 222 authority does not extend to cable broadband and mobile broadband internet providers Restoring Internet Freedom 83 Fed Reg 7852 (Feb 2 2020) (to be codified at 47 CFR pts 1 8 and 20) This regulatory shift was subsequently upheld by the DC Circuit Mozilla Corporation v Federal Communications Commission 940 F3d 1 (2019)
In addition to sect 222 the FCC has authority to regulate ldquocommon carriersrdquomdashincluding both telecommunication carriers and broadband internet providersmdashunder sect 201(b) of the Communications Act In the past the FCC has interpreted sect 201(b) to protect against ldquounjust and unreasonablerdquo privacy and data security practices with respect to custom-ersrsquo personal information beyond CPNI In 2016 the FCC promulgated a regulation asserting its authority under this interpretation However Congress overturned this regulation pursuant to the Congressional Re-view Act in 2017 SJ Res 34 115th Cong (2017) At present the extent of the FCCrsquos authority under sect 201(b) remains unsettled (Mulligan and Linebaugh 2019)
Consumer Protection
The collection storage release and transmission of digital user data in-cluding proximity contacts is more generally governed by the Federal Trade Commission (FTC) The FTC is an independent US law enforce-ment agency tasked with protecting consumers and promoting competi-tion across broad sectors of the economy (FTC 2020) The FTCrsquos primary legal authority with respect to consumer protection comes from Section 5 of the FTC Act which prohibits ldquounfair or deceptive acts or practices in or affecting commercerdquo 15 USC sect 45(a)(1) Note that the FTC and FCC have some overlapping authority to protect consumer privacy in the context of telecommunications (FCC and FTC 2017)
The FTC has interpreted Section 5 to require companies to be trans-parent and accurate about their collection of PII from consumers A com-pany may be found to have engaged in a deceptive practice if it fails to disclose that it is collecting user data or fails to disclose that it is sharing these data with third parties and to provide a general description of these third parties The FTC has used its authority under Section 5 numerous
Legal Considerations 79
times to discipline companies that purport in published privacy policies or other notices to provide protection for the privacy andor security of personal information yet fail to do so in practice For example the FTC may find it both ldquounfairrdquo and ldquodeceptiverdquo for a mobile app privacy policy to state that the app never discloses location information to third parties when in fact the app shares that information with the app developerrsquos service provider which in turn uses it to provide analytical data to the app developer that are used to create targeted advertising
The FTC does not use its Section 5 authority other than to protect consumers and generally does not consider ldquode-identifiedrdquo user data which are data that are not ldquoreasonably linkablerdquo to a consumer to be a subject for consumer protection In general data collected are not ldquorea-sonably linkablerdquo so long as the company collecting it ldquo(1) takes rea-sonable measures to ensure that the data are de-identified (2) publicly commits not to try to reidentify the data and (3) contractually prohibits downstream recipients from trying to reidentify the datardquo (FTC 2012)
Many states have laws that are similar to Section 5 prohibiting un-fair and deceptive acts and practices Both Section 5 and these similar state laws can be violated not only by misrepresentations (affirmative deception) but also by material omissions Thus a failure to inform an app user of the apprsquos collection of tracking data and the planned use and disclosure of those data could constitute a violation of these laws Com-panies providing DCTT apps should make sure that all such information is disclosed in the appsrsquo terms of use to which users must affirmatively agree
Childrenrsquos Online Privacy
Children who use DCTT may be protected by additional privacy protec-tions In particular collection of digital PII from children under the age of 13 is strictly regulated under the Childrenrsquos Online Privacy Protection Act (COPPA) (15 USC sectsect 6501ndash6505) Under COPPA PII includes ldquofirst and last name[] a persistent identifier that can be used to recognize a user over time and across different online services[] and geolocation infor-mation sufficient to identify street name and name of a city or town[]rdquo COPPA prohibits a website or online service from collecting personal information (including location information) from children under age 13 without obtaining verifiable parental consent Note that there may be an
80 Digital Contact Tracing for Pandemic Response
exception to this requirement for an ldquoinvestigation on a matter related to public safetyrdquo 16 CFR sect 3125(c)(6)(iv)
Electronic Surveillance
In addition to misuse of user data by DCTT providers another privacy concern is that a third party may be able to access sensitive PII that is collected and stored by a DCTT system without the userrsquos knowledge and consent There are a number of federal criminal laws however that would likely prohibit such unauthorized access to PII
In particular the Electronic Communications Privacy Act of 1986 (ECPA)mdashwhich includes the Wiretap Act (18 USC sectsect 2510ndash2522) the Stored Communications Act (18 USC sectsect 2701ndash2711) and the Pen Register Act (18 USC sectsect 3121ndash3127)mdashmakes it a crime to access elec-tronic communications without authorization Individuals who violate the ECPA face up to five years in prison and fines up to $250000 Victims are also entitled to bring civil suits and recover actual damages in addi-tion to punitive damages and attorneyrsquos fees for violations
Generally the access restrictions in the ECPA apply unless consent is given or if access is authorized by statute for law enforcement purposes For example an employer is generally forbidden from accessing an em-ployeersquos private emails However if consent is given in the form of an employment contract that explicitly authorizes the employer to access emails it may be lawful under the ECPA for the employer to access such information Along the same lines the ECPA would likely prohibit an employer from accessing contact tracing data on an employeersquos phone without the employeersquos consent However the ECPA would likely not prohibit duly authorized government public health officials from access-ing contact tracing data without consent
As its name suggests the Stored Communications Act (SCA) regu-lates access to communications at rest that is not in transit The SCA makes it unlawful to intentionally access a facility in which electronic communication services are provided and to obtain alter or prevent au-thorized access to a wire or electronic communication while it is in elec-tronic storage in such a system As such the SCA would likely apply only to centralized collection of contact tracing data
The Pen Register Act covers any ldquosignaling informationrdquo exchanged in a communication such as phone numbers The statute does not reach
Legal Considerations 81
the content of such communications however An expansive interpreta-tion of the Pen Register Act would cover Bluetooth ldquohandshakesrdquo as they are merely signaling information between devices which do not carry content See United States v Forrester 512 F3d 500 (9th Cir 2007) (find-ing that email headers and IP addresses are akin to pen registers and have no Fourth Amendment protection) Unlike the SCA there is no statutory exclusionary rule that applies when the government illegally uses a pen register trap and trace device Additionally there is no private cause of action against the government for violations of the Pen Register Act
State Data Privacy Laws
States have a variety of privacy laws and are increasingly seeking to reg-ulate the online collection of personal information and the use and dis-closure of such information To date most of these laws focus more on transparency and protection from unauthorized access than on restricted collection and use (except with respect to biometric information) seek-ing to ensure that individuals who use websites or online services such as mobile applications do so on an informed basis with respect to the privacy provided by those sites and services Two examples of such state laws are the California Online Privacy Protection Act (CalOPPA) and the California Consumer Privacy Act (CCPA) Both laws require notice to in-dividuals who use websites or online services such as mobile applications in order to ensure that users are informed about the privacy of personal information collected by those sites and services (The CCPA also applies to data collection off-line) Both laws treat IP addresses and location data as types of potentially identifiable personal data and so would very likely apply to DCTT apps used by California residents
CalOPPA requires that the operator of any website mobile appli-cation or other online service (ldquoSiterdquo) post a privacy policy on the Site disclosing certain information regarding the Sitersquos collection use and dis-closure of PII CalOPPA applies to any Site that is accessible to California residents The required disclosures are not onerous and would apply only to collection of data that are identifiable to an individual person (but depending on who collects the data location data together with a device identifier are identifiable to the user)
The CCPA requires that any entity qualifying as a ldquobusinessrdquo provide its ldquoconsumersrdquomdashdefined as lawful residents of Californiamdashwith specific
82 Digital Contact Tracing for Pandemic Response
disclosures about the businessrsquos collection use and disclosure of personal information Importantly the CCPA applies only to for-profit businesses that meet certain thresholds of revenue or access to consumer informa-tion A public health agency or a nonprofit organization would not be subject to the CCPA Cal Civ Code sect 1798140(c)
The CCPA defines ldquopersonal informationrdquo as ldquoinformation that iden-tifies relates to describes is reasonably capable of being associated with or could reasonably be linked directly or indirectly with a particular consumer or householdrdquo The statute provides a nonexclusive list of po-tential identifiable personal information including ldquogeolocation datardquo In accordance with the CCPA businesses must provide consumers with a notice ldquoat or before the point of collectionrdquo of personal information which must describe the personal information to be collected and the pur-poses for collecting that information Businesses must additionally allow consumers to request access to and request deletion of personal informa-tion Businesses must allow for consumers to opt-out of the sale of any personal information Developers of COVID-tracing apps would want to build in compliance with these requirements In addition California Civil Code sect 1798815(a)(1) requires companies to ldquomaintain reason-able security procedures and practices appropriate to the nature of the information it processesrdquo
Like privacy laws generally the CCPA does not grant consumers rights regarding the use of de-identified information However the CCPA does require businesses to implement processes that prohibit re-identification of de-identified information as well as technical safeguards to prevent inadvertent release of that information Cal Civ Code sect 1798140(h)
Health Information Privacy
Many DCTT systems will be designed to collect health-related data of users such as symptom tracking SARS-CoV-2 test results and prior ex-posure to a person who is COV+ Individuals may have additional privacy protections with respect to the use and disclosure of this health-related information
The use and disclosure of individually identifiable health information is strictly regulated under the privacy and security rules implementing the
Legal Considerations 83
Health Insurance Portability and Accountability Act (HIPAA) HIPAA is limited in application however to health care providers and health insur-ance plans (ldquocovered entitiesrdquo) and ldquobusiness associatesrdquo of such entities ldquoBusiness associatesrdquo under HIPAA are persons who perform services for covered entities and need access to personal health information to do so
HIPAA-covered entities must have written authorization to use or disclose identifiable health information (ldquoprotected health informationrdquo or PHI) from the individual to whom such information pertains unless the HIPAA regulations promulgated by the US Department of Health and Human Services (HHS) provide an exception to the requirement for such individual authorization
Among the exceptions to the individual authorization requirement is an exception for certain uses and disclosures of PHI for public health purposes 45 CFR sect 164512(b) This exception would permit for exam-ple a HIPAA-covered entity to disclose the PHI of an individual who tests positive for SARS-CoV-2 to a public health authority A ldquopublic health authorityrdquo is an agency or authority of the US government a state ter-ritory a political subdivision of a state or territory or Indian tribe that is responsible for public health matters as part of its official mandate as well as a person or entity acting under a grant of authority from or under a contract with a public health agency such as a contact tracer Id sect 164501
Many DCTT developers are HIPAA business associates and any use and disclosure of PHI collected through DCTT used on behalf of HI-PAA-covered entities is restricted under the HIPAA privacy rules Nota-bly in response to COVID-19 HHS announced that its Office for Civil Rights would exercise its enforcement discretion and would not impose penalties for violations of certain provisions of the HIPAA Privacy Rule against health care providers or their business associates for the good faith uses and disclosures of protected health information for public health and health oversight activities during the nationwide public health emergency 85 FR 19392 (2020)
Many states also have health information privacy laws The HIPAA privacy rule sets a ldquofloorrdquo of privacy protections allowing the states to be more protective of privacy More specifically HIPAA preempts a state law if (but only if) the state law is ldquocontraryrdquo and less protective of privacy than the HIPAA privacy rule However if a state law is determined by the
84 Digital Contact Tracing for Pandemic Response
Secretary of HSS to be necessary to serve a ldquocompelling need related to public health safety or welfarerdquo it may survive preemption even if it is less privacy-protective than HIPAA 45 CFR sect 160203 (a)(1)(iv)
The Public Health Service Act also restricts the use of certain per-sonally identifiable information collected by entities involved with public health activities without the individualrsquos consent 42 USC 242m(d)
Labor and Employment Privacy Rights
Labor and employment lawsmdashthat is laws that govern the relationships between employers and employeesmdashmay prove relevant to DCTT espe-cially if employers mandate the use of DCTT or seek to collect health information regarding their employees using DCTT Depending on the built-in privacy protections of the DCTT system an employer may be able to access important health information from an employeersquos phone As noted above the ECPA would generally prohibit an employer from ac-cessing this information without the employeersquos consent Even with con-sent however there are limits on the collection and use of an employeersquos health information
In particular the use of DCTT may raise special concerns about em-ployment discrimination for example if an employer were to fire an em-ployee who tests positive for SARS-CoV-2 (COV+) or who has a known SARS-CoV-2 exposure The Americans with Disabilities Act (ADA) pro-tects disabled employees from discrimination and restricts the collection of personal health information by employers The Equal Employment Opportunity Commission (EEOC) which is the federal agency tasked with enforcing the ADA in the employment context would likely con-sider COV+ to be a ldquodisabilityrdquo under the ADA and analogous state laws prohibiting discrimination against disabled people COV+ is likely to be a ldquodisabilityrdquo especially where the individual is symptomatic andor experi-ences related health issues or if it is later determined that testing positive for SARS-CoV-2 leads to long-term or chronic health effects ldquoExposure to a COV+ personrdquo could also be covered by those laws because a person exposed to a COV+ individual could well be perceived as being disabled by being considered likely to be infected
Legal Considerations 85
The ADA generally requires that businesses make ldquoreasonable accom-modationsrdquo for persons who are disabled which may include individuals who are COV+ or who have a preexisting disability that places them at higher risk from or may be exacerbated by COVID-19 The EEOC has published guidance on reasonable accommodations under the ADA and related laws in the context of COVID-19 (EEOC 2020) Among other things this guidance clarifies that consistent with the ADA employers may take temperatures or otherwise collect health information about employees during the pandemic crisis so long as they keep that infor-mation confidential As of May 18 2020 the EEOC has not provided guidance that specifically addresses the applicability of the ADA to the use of DCTT by employers
In addition employment laws such as the ADA and the Family and Medical Leave Act (FMLA) and state law equivalents generally limit disclosure of information and require employers to keep confidential any employee personal health information related to a disability or request for medical leave Under the ADA any information regarding the medical condition or history of an employee that an employer obtains as part of an examination or inquiry into a disability could constitute a confidential medical record that can be disclosed only to certain individuals in lim-ited circumstances 42 USC sectsect 12112(d)(3)(B) and 12112(d)(4) The FMLA also prevents the disclosure of records related to medical histories in connection with an employeersquos leave request or eligibility 29 CFR sect 825500(g) The EEOC and some courts have gone further and taken the position that any information concerning an employeersquos medical con-dition is protected under the ADA or FMLA
As discussed elsewhere in this guidance document employers may have a good reason to employ DCTT in order to ensure workplace safety and limit the spread of SARS-CoV-2 in the community Employers may also face legal liability if they fail to protect employees (or customers) from potential exposure or infection In particular employers have an obligation under the Occupational Safety and Health Act to keep the workplace safe for employees In response to COVID-19 the Occupa-tional Safety and Health Administration (OSHA) has developed guid-ance on preparing the workplace (OSHA 2020) The CDC has also pre-pared guidance on healthy business operations and reducing the spread
86 Digital Contact Tracing for Pandemic Response
of SARS-CoV-2 in the workplace (CDC 2020c) Employers must strike an appropriate balance between avoiding employment discrimination and promoting workplace safety
Reflecting the need for such a balance the employee protections un-der the ADA and other employment laws are not absolute and are limited by among other things the need to protect the health and safety of other employees and the public Protection for workplace safety and health generally will justify appropriately tailored measures such as inquiries into an employeersquos personal health status or whether someone has tested positive for SARS-CoV-2 temperature checks and removal of employees from the workplace who are experiencing symptoms or have tested posi-tive and have not been cleared to return to work
Note finally that the use of DCTT by employers should be evaluated in conjunction with the hazard pay sick leave and other benefits that are available to employees Under the Families First Coronavirus Response Act employers with more than 50 employees and fewer than 500 employ-ees are required to provide two weeks of paid sick leave to an employee who stays home because of COVID-19 Pub L No 116-127 134 Stat 178 (2020) This paid leave extends to those who are themselves ill are quarantined or are awaiting a diagnosis as well as those who are caring for sick family members However reporting suggests that more than 75 of US workers will not qualify for benefits under this act (Cochrane Miller and Tankersley 2020)
Constitutional Privacy Rights
A DCTT program involving only private actors operating on the ba-sis of voluntarily provided information would not present constitutional privacy issues But any government-directed use of digital technology to support public health tracking and contact tracing involving mandatory government surveillance may potentially implicate a variety of consti-tutional protections These constitutional protections apply to actions taken by any level of government in the United States While state gov-ernments have broad policing powers in the area of public health (Jacob-son v Massachusetts 197 US 11 (1905)) and are generally allowed to enforce legislation not preempted by federal laws even emergency and
Legal Considerations 87
health-protective laws must be consistent with the US Constitution (HHS 2019 CDC 2020f)
Fourth Amendment Search and Seizure
Many people considering whether to use a DCTT app may be concerned that government enforcement agencies would obtain tracing data and use those data to conduct criminal prosecutions or immigration proceed-ings Constitutional protections notably the Fourth Amendmentrsquos limit on warrantless searches limit the governmentrsquos use of personal data in the criminal context However exceptions exist allowing law enforce-ment to access information even when such access would generally be prohibited How the government accesses personal data stemming from contact tracing needs to be scrutinized and protections will hinge on the manner of access
In general the Fourth Amendment protects ldquo[t]he right of the peo-ple to be secure in their persons houses papers and effects against un-reasonable searches and seizuresrdquo As originally interpreted the Fourth Amendment was considered tied to common-law trespass That is no lon-ger the case US Supreme Court precedent interprets the Fourth Amend-ment to protect ldquopeople not placesrdquo and extends to the protection of certain expectations of privacy such as location information as long as such expectations are reasonable Katz v United States 389 US 347 351 (1967) A warrantless government search is unconstitutional when the information sought is private and such expectation of privacy is ldquoone that society is prepared to recognize as reasonablerdquo Smith v Maryland 442 US 735 743ndash44 (1979)
The constitutionality of a search will revolve around the following analysis whether the digital program either violates an individualrsquos ldquorea-sonable expectation of privacyrdquo (likely triggered by programs collecting large amounts of location andor health data) or involves a government ldquotrespassrdquo (likely triggered by required app downloads) Katz v United States 389 US 347 (1967) United States v Jones 565 US 400 (2012)
Courts will most likely weigh the intrusiveness of the measures taken in implementing a search standard against the severity of the situation governmental and individual interests and accountability measures and safeguards built into the system
Voluntary sharing by individuals of their information with other par-
88 Digital Contact Tracing for Pandemic Response
ties including the government would mean that there was no reasonable expectation of privacy and would not raise the issues elaborated above It is worth noting that consent may not be considered voluntary if coerced or conditioned especially with regard to public employees or students of public institutions
Third-Party Doctrine
Some legal doctrines allow for the governmentrsquos acquisition of otherwise private information consistent with Fourth Amendment privacy protec-tions The third-party doctrine for example provides that individuals have no reasonable expectation of privacy in information voluntarily shared with others even if the information is revealed on the assumption that it will be used only for a limited purpose and the confidence placed in the third party will not be betrayed Smith v Maryland 442 US 735 (1979) United States v Miller 425 US 435 (1976) This applies to in-formation provided by third parties (mobile carriers internet service pro-viders medical tracking device manufacturers etc) to the government under order or request even when the third partyrsquos end-user agreements or privacy policies create an expectation of privacy
The Supreme Court has narrowed the applicability of the third-party doctrine to exclude use and disclosure of ldquohistoricalrdquo cell-site location information (CSLI) data For example in Carpenter v United States 138 S Ct 2206 (2018) the Court reasoned that the third-party doctrine does not justify use and disclosure of historical CSLI because an individual does not provide that information voluntarily Rather that information is pervasively collected by the cell phone company without any affirmative action on the part of the individual The Court did not express a view on ldquoreal-timerdquo CSLImdashlocation information that live-tracks a cell phonersquos locationmdashor on GPS data that may be stored in the phone itself
The Carpenter decision builds on a line of cases related to searches of digitally stored location data In Riley v California 134 S Ct 2473 (2014) the Court held that absent exigent circumstances law enforce-ment must obtain a warrant to search an individualrsquos phone Exigent circumstances are those that require immediate action because there is a probability that evidence may be destroyed The use of a centralized data-base for collection of digital contact tracing data would obviate deletion
Legal Considerations 89
concerns If the data are stored locally in the phone issues may arise as to whether law enforcement may suspect the data may be deleted following an arrest
Similarly in United States v Jones 132 S Ct 945 (2012) Justice So-nia Sotomayor authored a concurring opinion arguing that the use of a GPS to track a defendantrsquos whereabouts has the potential of providing the government with enough data points to create a ldquomosaicrdquo of the personrsquos life Location data obtained through centralized location contact tracing have the potential of providing information on an individualrsquos where-abouts beyond whatrsquos necessary for determining proximity to infected individuals Localized data may also raise the same issues if accessed by law enforcement
Following Carpenter several courts have addressed the constitution-ality of novel location tracking In Massachusetts for instance a federal district court concluded that police use of a ldquopole camerardquo on a utility pole to investigate the movements of an individual constituted a search under the Fourth Amendment United States v Moore-Bush 381 FSupp3d 139 (D Mass 2019) The court reasoned that even in a public space an in-dividual still retains a reasonable expectation of privacy ldquoin the whole of their physical movementsrdquo Citing Carpenter and Jones the court stated that the governmentrsquos unrestrained power to collect data that reveal pri-vate aspects of identity is susceptible to abuse and gives police access to a category of information that is ldquootherwise unknowablerdquo Long-term monitoring of a personrsquos movements consequently violates that individ-ualrsquos expectation of privacy Notably the court emphasized the capability of the camera to create a searchable digital log of the photos taken for the eight-month period during which the camera was used
State courts have also weighed in on the issue The Massachusetts Supreme Judicial Court found that police access to real-time location data pinpointing an individualrsquos movement whether from a third party or a cell-site simulator infringes upon an individualrsquos reasonable expec-tation of privacy Commonwealth v Almonor 120 NE3d 1183 1195 (Mass 2019) The Washington Supreme Court for its part held that a cell phone ping used to locate the defendantrsquos vehicle in real time is a search under the Fourth Amendment requiring a warrant absent exigent circumstances State v Muhammad 428 P3d 1177 (2018) And the Colo-
90 Digital Contact Tracing for Pandemic Response
rado Court of Appeals held that police use of a video pole camera to con-tinuously surveil a defendantrsquos fenced-in backyard constitutes a search under the Fourth Amendment People v Tafoya 2019 BL 457321 Colo Ct App 17CA1243 (2019)
Application of Carpenter by lower courts to novel location-tracking tactics is still evolving and it is as yet unclear how the narrower interpre-tation of the third-party doctrine will continue to be expanded and ap-plied particularly in cases of short-term monitoring of massive amounts of location andor health data Moreover it is unclear whether Carpenter would apply to DCTT data collected by the government itself
Special Needs Doctrine
An argument in favor of the constitutionality of government DCTT programs is that the ldquospecial needsrdquo doctrine would apply Under this doctrine a warrantless search that would otherwise violate the Fourth Amendment might be permissible based on a special need relating to pub-lic health When the search is conducted for a nontraditional law enforce-ment purpose and circumstances make securing a warrant impracticable the Supreme Court has ruled that warrantless searches may be permissi-ble The special needs doctrine however is highly controversial because it is not a consistently applied Fourth Amendment exception so it is diffi-cult to predict when courts would authorize nontraditional surveillance Some factors considered by the court are (1) the balance between the intrusiveness of the government action and the anticipated public bene-fits (2) the existence of legislative authorization (3) judicial process or the ability of the subject individual to challenge the government action (4) the scope or breadth of government action and (5) the likelihood of the collected data being used in criminal proceedings The Supreme Court did note in Chandler v Miller 520 US 305 (1997) that a ldquorisk to public safety [that] is substantial and realrdquo may justify ldquoblanket suspicionless searches calibrated to the riskrdquo citing as examples the routine searches conducted at airports and entrances to some official buildings (Searches within the context of immigration are further analyzed below)
Immigration Enforcement
Exceptions apply to the constitutional requirement that a warrant ac-company an unreasonable search or seizure in the immigration context
Legal Considerations 91
For example an exception to the general warrant requirement is the bor-der search exception which allows government officials to search and seize without a warrant persons and property at the border or at the functional equivalent of a border See United States v Montoya de Her-nandez 473 US 531 (1985) United States v Flores-Montano 541 US 149 (2004) Federal regulation authorizes immigration officials to oper-ate within 100 miles of any US external boundary (See 8 CFR sect 2871 defining ldquoreasonable distancerdquo as ldquowithin 100 air miles from any external boundary of the United Statesrdquo) A functional equivalent of a border may include any airport where international flights may be received automo-bile checkpoints servicing international traffic and vessels in territorial waters Government officials however must still have ldquoreasonable suspi-cionrdquo of an immigration violation or a crime to search or seize persons or property
In the context of digital data Customs and Border Protection (CBP) officials may conduct either manual or forensic searches of electronic devices at the border or its functional equivalent A manual search is considered a routine search and may include accessing the phone and ldquobrowsingrdquo its contents If the electronic device is password protected individuals must provide information for unlocking the device Forensic searches on the other hand are nonroutine and involve a more invasive search of the electronic devicersquos contents Federal circuit courts are split on whether a CBP agent needs ldquoreasonable suspicionrdquo before conducting a forensic search of an electronic device But Supreme Court precedent clearly states that suspicionless searches are not unconstitutional when public safety is considered Skinner v Ry Labor Execsrsquo Assrsquon 489 US 602 (1989)
A recent CBP directive provides guidance and standard operating procedures regarding forensic searches of electronic devices CBP 3340-049A Border Search of Elec Devices (DHS 2018) The directive states that CBP officers may detain electronic devices or copies of the informa-tion contained within these devices for a reasonable period time not to exceed five days This directive raises the concern that travelers may be required to turn over contact tracing data stored on their phone to CBP officers Note that the directive has been challenged in federal court and is currently awaiting appeal Alasaad v Nielsen 419 FSupp3d 142 (D Mass 2019)
92 Digital Contact Tracing for Pandemic Response
Searches in Schools
Another exception to the general warrant requirement applies to searches by non-law-enforcement government officials in public schools (ie school officials) Within this context school officials have broad powers to conduct searches as long as those searches are reasonable Searches by individuals in private schools are not governed by the Fourth Amend-ment State regulation of searches in private schools varies (See US DOE 2009)
Related Federal Privacy Statutes
Outside the Fourth Amendment context certain laws provide protections against government collection of and access to personal data The USA Freedom Act of 2015 for example bans the governmentrsquos bulk collec-tion of internet metadata and telephonic records which was previously allowed under Section 215 of the USA Patriot Act The government must now identify with specificity the identity of a person account address or personal device when requesting records The law allows for the acqui-sition of data by two degrees of separationmdashor ldquohopsrdquomdashfrom targeted individuals If a centralized system in contact tracing is used it is unclear whether the government may need to resort to this provision since it would likely have consent from individuals to collect and use the data
The Privacy Act of 1974 also regulates the collection use and disclo-sure of personal data but applies only to federal agencies (and their con-tractors) not to state or local agencies 5 USC sect 552a The Act protects against disclosure of individually identifying ldquorecord[s]rdquo that are kept within a ldquosystem of recordsrdquo The Act limits disclosure of information ldquoexcept pursuant to a written request by or with prior written consent of the individual to whom the record pertainsrdquo Certain disclosures are ex-empt from the Actrsquos applicability Pertinent disclosure exceptions are for records required to be disclosed under the Freedom of Information Act (FOIA) or disclosures ldquoto a person pursuant to a showing of compelling circumstances affecting the health or safety of an individualrdquo A disclosure under FOIA however would not include information in ldquopersonnel and medical files and similar filesrdquo when disclosure ldquowould constitute a clearly unwarranted invasion of personal privacyrdquo FOIA Guide 2004 Edition
Legal Considerations 93
Exemption 6 If non-anonymized data are turned over to the federal or state governments it is important to consider whether PII would be pro-tected from disclosure under FOIA or state freedom of information laws
Consent
User consent is a cross-cutting issue for evaluating many of the laws and regulations governing personal information privacy discussed in the prior sections In general privacy laws can be justified on the grounds that an individual should have the option to control with various types and de-grees of limitation the collection use retention andor disclosure of in-formation pertaining to that individual by others As such many privacy laws start from the premise that absent an individualrsquos consent use or disclosure of that individualrsquos PII is impermissible except for certain enu-merated purposes deemed to outweigh the individualrsquos privacy interests
Consent like agreements in general can be manifest in different ways in specific circumstances In some cases an affirmative actionmdashsuch as a signaturemdashis needed to demonstrate consent In other cases inactionmdashsuch as declining to ldquounsubscriberdquo from receiving certain unsolicited emailsmdashconstitutes consent Where a law requires a ldquowrittenrdquo signature in all but a few contexts the signature may be executed electronically In the United States that means the ldquosignaturerdquo may consist of any of the following ldquoan electronic sound symbol or processrdquo so long as it is ldquoattached to or logically associated with a contract or other record and executed or adopted by a person with the intent to sign the recordrdquo Electronic Signatures in Global and National Commerce Act (E-SIGN) (15 USC 7006)
The scope of a consent depends on what was deemed to be under-stood by the consenting party That is least clear when the consent is in-ferred from inactivity even if terms stating the consequences of inactivity have been provided The scope of consent is most clear when the terms agreed to have been presented to or provided by the consenting party in a conspicuous documented manner and a record exists of those terms Courts uphold the validity of clickthrough agreements because users are deemed to review the terms to which they respond by clicking ldquoI agreerdquo
94 Digital Contact Tracing for Pandemic Response
But where terms are ambiguous or confusing buried in other text or presented obscurely the ldquoI agreerdquo action may not mean the user actually agreed to specific terms
Terms of Use and Privacy Policies for apps are often written in com-plicated or nuanced language with key points difficult to discern More-over they are generally hard to read on a mobile device Many users of mobile phone apps agree to such terms without even attempting to read or to understand them As such it is often questionable whether an app user has knowingly agreed to all the terms of those documents Presen-tation of terms in large typeface short sentences simple language and direct disclosures makes user consent more meaningful
For contact tracing apps that collect PII andor PHI consent will overcome the restrictions of many if not most privacy laws provided the consent is freely given reflects a full understanding of the terms for use collection and disclosure of the information and is confirmed by an af-firmative act such as a click that may be executed only upon a complete reading of Terms of Use and Privacy Policies Whether consent may be deemed ldquofreely givenrdquo in certain circumstances depends on contextual understandings of party relationships including the employer-employee and government-citizen relationships
Anti-discrimination and Individual Freedom Laws
Any measure taken to protect public health and safety must comply with the Constitution and civil rights laws such as the ADA that prohibit discrimination against persons in certain protected categories such as race gender religion or disability In addition certain implementations of DCTT could be challenged under a variety of individual freedom protections
Anti-discrimination Laws
In general it would be impermissible to use DCTT in a way that either targets or excludes people on the basis of their membership in one of these protected categories
When motivated by animus against a protected class as defined by law and not narrowly tailored to advance a compelling government inter-
Legal Considerations 95
est a discriminatory regulation would be considered unconstitutional un-der the Equal Protection Clause of the Fourteenth Amendment to the US Constitution See Jew Ho v Williamson 103 F10 CCND Cal (1900) (striking down a quarantine imposed by San Francisco in response to an outbreak of bubonic plague because it was racially motivated) see also Church of Lukumi Babalu Aye v Hialeah 508 US 520 (1993) (supposed public health measure unconstitutional because it targeted the practices of one religion)
The risk of unintentional yet illegal discrimination in using DCTT is a real possibility Recent studies of infection rates among the population have revealed a larger-than-proportional infection rate among certain minority communities such as Latinx African American and American Indian communities (NYC DOH 2020) Programs that target specific ra-cial ethnic tribal or religious groups may raise constitutional and other legal concerns
Religious Freedom Laws
The use of DCTT may also raise concerns about religious freedom For example there may be religious objections to restrictions on gathering for worship carrying a mobile phone or the use of imaging technology Under current Supreme Court precedent generally applicable laws that do not discriminate against religion on their face do not violate the Free Exercise Clause of the First Amendment even if those laws have an inci-dental effect on the exercise of religion Employment Div Dept of Hu-man Resources of Oregon v Smith 494 US 872 (1990) These laws need not be justified by compelling government interest (the ldquostrict scrutinyrdquo standard of review) the government need only show that they are ratio-nally related to a legitimate interest On the other hand laws that are not neutral and not of general applicability must be justified by compelling government interest and must be narrowly tailored to advance that in-terest if it burdens religious practicesmdasha very tough hurdle to overcome Church of Lukumi Babalu Aye v Hialeah 508 US 520 (1993)
This general approach however is disrupted in some contexts by statutes adopted to provide greater protection to religious freedom The federal Religious Freedom Restoration Act (RFRA) requires strict scru-tiny for federal actions that burden religion and many states have ad-opted ldquostate RFRAsrdquo that do the same for actions by state and local
96 Digital Contact Tracing for Pandemic Response
governments The Religious Land Use and Institutionalized Persons Act which extends similar protections to persons confined to an institution such as a prison jail or mental health facility may also be relevant 42 USC sect 2000cc
Under either standard of review courts will examine whether a gov-ernment action imposes a substantial burden on religious exercise if not no religious freedom violation has occurred Such a finding is unlikely for DCTT programs absent evidence that the government is using the digital information to take action against religious persons that is not necessary to avoid the spread of a serious disease Nevertheless legal challenges on religious freedom grounds cannot be ruled out
Legislative Recommendations
bull Congress should enact new legislation specifically tailored to fa-cilitate the use of DCTT as part of the public health response to COVID-19 while also protecting user privacy and ensuring data security
bull Congress should require DCTT developers to disclose to users in clear language the nature of the information that would be col-lected how it would be collected how it would be stored and for what purposes it may be used
bull While the rollout of DCTT should initially employ an opt-in au-thorization approach the feasibility acceptability and value of opt-out approaches should continue to be evaluated As such opt-out approaches to consent should not be precluded by legislation
bull Congress should prohibit the commercial use of data collected for COVID-19 response by DCTT
bull Congress should prohibit discrimination on the basis of data col-lected by DCTT
bull If Congress is unable to enact suitable legislation state legislatures should work toward enacting similar laws for their jurisdictions A ldquomodelrdquo state law should be rapidly developed to facilitate na-tionwide uniformity of legal requirements
97
Public Health
Characteristics That Make Data Useful to
Public Health for Reducing Disease Transmission
bull Technologies or apps with the goal of enhancing public health ca-pacity to identify cases and trace contacts in order to control the spread of SARS-CoV-2 should be designed to match functionality with that goal
bull Technologies or apps may produce some false negatives or false positives but they should be accurate enough that public health authorities feel confident that they support and donrsquot detract from contact tracing efforts
bull DCTT approaches for public health should be designed to facili-tate the following
deg Identifying contacts including those who may not be easily found otherwise
deg Finding and notifying contacts rapidly before they develop symptoms if infected
deg Analyzing the nature of contact to determine whether con-tact is high medium or low risk to support decisions about whether quarantine should be mandatory should be volun-tary or is not needed
deg Following up with cases and contacts so that public health can provide resources to support isolation and quarantine at home
Recommendations
FIVE
98 Digital Contact Tracing for Pandemic Response
bull Data collected through DCTT should be made available to public health professionals and to researchers in de-identified form to support population-level epidemiologic analysis
Ethics
Collecting Data to Inform Policy and Practice
bull Reviews of DCTT systems must be conducted in part by an inde-pendent intermediary that has established the publicrsquos trust
bull Those who authorize use of DCTT within a particular jurisdic-tion or institution should continuously and systematically monitor the technologyrsquos performance in that context This should include monitoring for effectiveness and benefit monitoring for harms and monitoring for the fair distribution of both benefits and harms They should also monitor evidence that is being generated in other contexts about their selected technological solution and about other competing technologies
bull Data should be available to users that would permit them to further investigate their personal risk with public health officials or other health workers to add a layer of protection against unnecessary quarantine
Public Trust and Public Attitudes
bull More research into public attitudes is needed In particular in-depth qualitative research should examine public attitudes about perceptions of trust in DCTT among different communities which features of DCTT influence trust and the extent to which people are willing to provide different types of data through DCTT to help their community
bull States and localities that are considering adopting DCTT should engage with the public to increase their understanding of the acceptability of DCTT design features and uses among diverse communities
Recommendations 99
Designing Flexible Technology to Maximize
Public Health Utility While Respecting Other Values
bull Technology companies should not alone control the terms condi-tions or capabilities of DCTT nor should they presume to know what may be acceptable to members of the public
bull A ldquovalues in designrdquo approach to development of DCTT should be adopted (Flanagan Howe and Nissenbaum 2008 Knobel and Bowker 2011) Robust public- and user-engagement activities should be pursued to identify and incorporate to the extent pos-sible a range of values into the design of the technology These values may include privacy but also autonomy efficiency equity or others Technology design should reflect an appropriate balance and prioritization of identified values
bull Technology design should not be static but should be capable of evolving depending upon local conditions new evidence and changing preferences and priorities
bull DCTT should be designed to have a base set of features that pro-tect privacy with layers of additional capabilities that users may choose to activate An initial default should be that user location data are not shared but users should be provided with easy mech-anisms and prompts to allow for opting-in to this capability with encouragement to the public if and as it is shown to be critical to achieving public health goals
Policy Positions to Advance Widespread
Use of Digital Contact Tracing Technologies
bull DCTT use should not be mandated at this time given uncertainty about potential harms and benefits Additional technology user and real-world testing is needed
bull Incentives can be a useful complement to encouragements how-ever any incentives for users to install and use DCTT must be equitable should not be coercive and should align with effective use of the technology
100 Digital Contact Tracing for Pandemic Response
bull Trusted leaders should be enlisted to communicate effectively with the public about DCTT and encourage its use should the technol-ogy demonstrate some potential The limits of knowledge regard-ing effectiveness should also be explained along with what will be done to improve technological capabilities as understanding evolves
Disclosure and AuthorizationConsent
bull A clear and concise module consisting of basic disclosure and vol-untary authorization should be developed to accompany DCTT This module should not take the form of ldquoclickwraprdquo terms of service or end-user agreements but rather provide only essential information necessary for an individual to make a decision More detailed disclosures (such as FAQs in plain language) should be made easily accessible to those who wish to learn more with no hidden surprises
bull An opt-in approach to authorization should be instituted to ac-company initial DCTT rollout The feasibility and value of opt-out approaches should continue to be evaluated informed by what is technologically possible what local assessments of benefits and harms of the technology reveal over time and our evolving un-derstanding of the degree to which an opt-out approach is likely to increase or decrease utilization among different populations Opt-out approaches should not be precluded
Promoting Equity and Fairness
bull A commitment to equity means a commitment to ensuring that the benefits and burdens of DCTT are distributed fairly Public engagement is an important tool for assessing impact and to rectify inequities
bull States localities and institutions that recommend widespread use of DCTT should provide technology (eg mobile phones Blue-tooth devices) and free data packages to those who desire but lack access to these devices
Recommendations 101
bull If there are lower rates of adoption of DCTT systems in some identifiable communities public health authorities should iden-tify ways to compensate For example directing more non-DCTT resources and efforts toward those communities to meet specific needs that are elsewhere being supported by technology
bull If maps are generated based on DCTT to provide the public with the locations that COV+ individuals have visited steps must be taken to minimize the stigma and potential financial losses that could result from being identified as a hotspot
Instituting Transparent Governance and Oversight
bull Digital surveillance oversight committees should be established expeditiously with diverse and qualified membership to provide ethical and regulatory review prior to and concurrent with wide-spread use of a DCTT system
bull Understandable and publicly accessible rules must guide the col-lection access control use storage and combination of data by government authorities public and private institutions and other parties such as public health researchers
bull Only those data that are necessary and relevant for the public health response to COVID-19 should be collected and used
bull Identifiable data should be kept only for the period of time needed for the public health response to COVID-19
bull Identifiable data collected as part of this response should not be shared with anyone other than the relevant public health authorities without additional specific informed consent of individual users
bull Before a government or institution adopts a digital contact tracing program they should state the conditions under which the digital contact tracing program will be terminated
bull Future use of DCTT to advance public health or other efforts (eg use in seasonal flu surveillance) would require independent justification DCTT designed for public health use should not be used by law or immigration enforcement
102 Digital Contact Tracing for Pandemic Response
bull The principles offered in this guidance document apply both during and following the COVID-19 pandemic
Legislative
bull Congress should enact new legislation specifically tailored to fa-cilitate the use of DCTT as part of the public health response to COVID-19 while also protecting user privacy and ensuring data security
bull Congress should require DCTT developers to disclose to users in clear language the nature of the information that would be col-lected how it would be collected how it would be stored and for what purposes it may be used
bull While the rollout of DCTT should initially employ an opt-in au-thorization approach the feasibility acceptability and value of opt-out approaches should continue to be evaluated As such opt-out approaches to consent should not be precluded by legislation
bull Congress should prohibit the commercial use of data collected for COVID-19 response by DCTT
bull Congress should prohibit discrimination on the basis of data col-lected by DCTT
bull If Congress is unable to enact suitable legislation state legislatures should work toward enacting similar laws for their jurisdictions A ldquomodelrdquo state law should be rapidly developed to facilitate na-tionwide uniformity of legal requirements
Resources
103
US Government Response
White House
The White House and CDC 16 April 2020 ldquoOpening Up America Againrdquo The White House and the Centers for Disease Control and Prevention Available at httpswwwwhitehousegovopeningamerica
Congress
A Bill to Protect the Privacy of Consumersrsquo Personal Health Information
Proximity Data Device Data and Geolocation Data during the Corona-
virus Public Health Crisis S3663 116th Cong (2019-2020) Available at httpswwwcongressgovbill116th-congresssenate-bill3663
A Bill to Protect the Privacy of Health Information during a National
Health Emergency S3749 116th Cong (2019-2020) Available at httpswwwcongressgovbill116th-congresssenate-bill3749
To Protect the Privacy of Health Information during a National Health
Emergency HR6866 116th Cong (2019-2020) Available at httpswwwcongressgovbill116th-congresshouse-bill6866
Congressional Research Service
Foster Michael 16 April 2020 ldquoCOVID-19 Digital Surveillance and Privacy Fourth Amendment Considerationsrdquo Legal Sidebar LSB10449 Congres-sional Research Service httpscrsreportscongressgovproductpdfLSBLSB10449
Mulligan Stephen P and Chris D Linebaugh 25 March 2019 ldquoData Pro-tection Law An Overviewrdquo R45631 Congressional Research Service httpscrsreportscongressgovproductpdfRR45631
104 Resources
US Department of Health and Human Services (HHS) Centers for Disease Control and Prevention (CDC)
CDC 6 May 2020 ldquoInterim Guidance for Businesses and Employersrdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention Available at httpswwwcdcgovcoronavirus2019-ncov communityguidance-business-responsehtml
CDC 3 May 2020 ldquoDiscontinuation of Isolation for Persons with COVID-19 Not in Healthcare Settingsrdquo Interim Guidance Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention httpswwwcdc govcoronavirus2019-ncovhcpdisposition-in-home-patientshtml
CDC 30 April 2020 ldquoContact Tracingrdquo Get and Keep America Open Sup-porting States Tribes Localities and Territories Centers for Disease Control and Prevention httpswwwcdcgovcoronavirus2019-ncovphpopen-americacontact-tracinghtml
CDC 29 April 2020 ldquoInterim Guidelines for Collecting Handling and Testing Clinical Specimens from Persons for Coronavirus Disease 2019 (COVID-19)rdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention Available at httpswwwcdcgovcorona-virus2019-ncovlabguidelines-clinical-specimenshtml
CDC 29 April 2020 ldquoContact Tracing Part of a Multipronged Approach to Fight the COVID-19 Pandemicrdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention httpswwwcdcgov coronavirus2019-ncovphpprinciples-contact-tracinghtml
CDC 28 April 2020 ldquoPreliminary Criteria for the Evaluation of Digital Contact Tracing Tools for COVID-19rdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention Available at httpswwwcdc govcoronavirus2019-ncovdownloadsphpprelim-eval-criteria-digital- contact-tracingpdf
CDC 20 April 2020 ldquoDigital Contact Tracing Tools for COVID-19rdquo Corona- virus Disease 2019 (COVID-19) Centers for Disease Control and Preven-tion Available at httpswwwcdcgovcoronavirus2019-ncovdownloadsdigital-contact-tracingpdf
CDC 6 April 2020 ldquoOperational Considerations for the Identification of Healthcare Workers and Inpatients with Suspected COVID-19 in Non-US Healthcare Settingsrdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention Available at httpswwwcdcgovcoron virus2019-ncovhcpnon-us-settingsguidance-identify-hcw-patientshtml
HHS 2 April 2020 ldquoNotification of Enforcement Discretion under HIPAA to Allow Uses and Disclosures of Protected Health Information by Business
Resources 105
Associates for Public Health and Health Oversight Activities in Response to COVID-19rdquo 45 CFR Parts 160 and 164 Available at httpswwwhhsgovsitesdefaultfilesnotification-enforcement-discretion-hipaapdf
HHS 30 March 2020 Notification of Enforcement Discretion for Telehealth Remote Communications during the COVID-19 Nationwide Public Health Emergency US Department of Health and Human Services Office of Civil Rights Available at httpswwwhhsgovhipaafor-professionalsspecial- topicsemergency-preparednessnotification-enforcement-discretion- telehealthindexhtml
CDC 24 February 2020 ldquoLegal Authorities for Isolation and Quarantinerdquo Centers for Disease Control and Prevention Available at httpswww cdcgovquarantineaboutlawsregulationsquarantineisolationhtml
HHS 26 November 2019 ldquoPublic Health Emergency Declarationrdquo Public Health Emergency US Department of Health and Human Services Office of the Assistant Secretary for Preparedness and Response Available at httpswwwphegovPreparednesslegalPagesphedeclarationaspx
CDC September 2018 ldquoPublic Health Surveillance Preparing for the Futurerdquo Centers for Disease Control and Prevention Available at httpswww cdcgovsurveillancepdfsSurveillance-Series-Booklethpdf
CDC 2005 ldquoVI Plan for Surveillance of Contacts of SARS Case Supplement B SARS Surveillancerdquo Centers for Disease Control and Prevention Available at httpswwwcdcgovsarsguidanceb-surveillancecontactshtml
National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research 1979 ldquoThe Belmont Report Ethical Principles and Guidelines for the Protection of Human Subjects of Researchrdquo US Depart-ment of Health and Human Services Office of Human Research Protec-tions httpswwwhhsgovohrpregulations-and-policybelmont-reportread-the-belmont-reportindexhtml
US Equal Employment Opportunity Commission (EEOC)
EEOC 7 May 2020 ldquoWhat You Should Know About COVID-19 and the ADA the Rehabilitation Act and Other EEO Lawsrdquo US Equal Employ-ment Opportunity Commission Available at httpswwweeocgovwyskwhat-you-should-know-about-covid-19-and-ada-rehabilitation-act-and-other-eeo-laws
EEOC 21 March 2020 ldquoPandemic Preparedness in the Workplace and the Americans with Disabilities Actrdquo US Equal Employment Oppor-tunity Commission Available at httpswwweeocgovlawsguidancepandemic-preparedness-workplace-and-americans-disabilities-act
106 Resources
Federal Trade Commission (FTC)
FTC 2020 ldquoPrivacy amp Data Security Update 2019rdquo Federal Trade Commis-sion httpswwwftcgovsystemfilesdocumentsreportsprivacy-data- security-update-20192019-privacy-data-security-report-508pdf
FTC 2012 ldquoProtecting Consumer Privacy in an Era of Rapid Changerdquo FTC Report Recommendations for Businesses and Policymakers Fed-eral Trade Commission httpswwwftcgovsitesdefaultfilesdocumentsreportsfederal-trade-commission-report-protecting-consumer-privacy-era- rapid-change-recommendations120326privacyreportpdf
Occupational Safety and Health Administration (OSHA)
OSHA April 2020 ldquoGuidance on Preparing Workplaces for COVID-19rdquo OSHA 3990-03 2020 Department of Labor Occupational Safety and Health Act of 1970 Available at httpswwwoshagovPublicationsOSHA3990pdf
Customs and Border Protection (CBP)
CBP 4 January 2018 ldquoBorder Search of Electronic Mediardquo US Customs and Border Protection CDP DIRECTIVE NO 3340-049A Available at httpswwwcbpgovsitesdefaultfilesassetsdocuments2018-JanCBP-Directive-3340-049A-Border-Search-of-Electronic-Media-Compliantpdf
Other Governmental and Nongovernmental Organizations
Contact Tracing Surveillance PlansMethods
Simmons-Duffin Selena 7 May 2020 ldquoStates Nearly Doubled Plans for Contact Tracers since NPR Surveyed Them 10 Days Agordquo NPR Available at httpswwwnprorgsectionshealth-shots20200428846736937we-asked-all-50-states-about-their-contact-tracing-capacity-heres-what- we-learned
Resolve to Save Lives 2020 ldquoCOVID-19 Contact Tracing Playbookrdquo Vital Strategies Available at httpscontacttracingplaybookresolvetosavelives org
Simpson Erin and Adam Conner 22 April 2020 ldquoDigital Contact Tracing to Contain the Coronavirusrdquo Center for American Progress Avail-able at httpswwwamericanprogressorgissuestechnology-policynews20200422483521digital-contact-tracing-contain-coronavirus
Watson Crystal Anita Cicero James Blumenstock and Michael Fraser 10 April 2020 ldquoA National Plan to Enable Comprehensive COVID-19 Case Finding and Contact Tracing in the USrdquo Johns Hopkins Bloomberg School of Pub-
Resources 107
lic Health Center for Health Security and the Association of State and Ter-ritorial Health Officials Available at httpswwwcenterforhealthsecurity orgour-workpubs_archivepubs-pdfs2020200410-national-plan-to- contact-tracingpdf
Africa CDC 9 April 2020 ldquoGuidance on Contact Tracing for COVID-19 Pandemicrdquo Manuals Guidelines amp Frameworks African Union Africa CDC Available at httpsafricacdcorgdownloadguidance-on-contact- tracing-for-covid-19-pandemic
PIH 4 April 2020 ldquoPart I Testing Contact Tracing and Community Manage-ment of COVID-19rdquo PIH Guide | COVID-19 STOP COVID Partners in Health Available at httpswwwpihorgsitesdefaultfiles2020-04PIH_Guide_COVID_Part_I_Testing_Tracing_Community_Managment_4_4pdf
Hellewell Joel Sam Abbott Amy Gimma et al 28 February 2020 ldquoFea- sibility of Controlling COVID-19 Outbreaks by Isolation of Cases and Contactsrdquo The Lancet Global Health 8 e488ndash96 Available at https doiorg101016S2214-109X(20)30074-7
Contact Tracing EthicsPrinciplesGuidance
Center for Democracy amp Technology (CDT)ndashhttpscdtorginsightsCDT 30 April 2020 ldquoStatement of the Center for Democracy amp Technol-
ogy Regarding Use of Data to Fight COVID-19rdquo Center for Democ-racy amp Technology httpscdtorgwp-contentuploads202004 CDT-Statement-Regarding-Use-of-Data-to-Fight-COVID-19pdf
Shetty Ridhi 23 April 2020 ldquoData Use in the Fight against COVID-19 Should Treat People Equitably Not Exacerbate Long-Standing Dispar-itiesrdquo Center for Democracy amp Technology httpscdtorginsightsdata-use-in-the-fight-against-covid-19-should-treat-people-equitably-not-exacerbate-long-standing-disparities
Greenwood Dazza Gregory Nadeau Pagona Tsormpatzoudi Bryan Wilson Jeffrey Saviano and Alex ldquoSandyrdquo Pentland 30 April 2020 ldquoCOVID-19 Contact Tracing Privacy Principlesrdquo MIT Computational Law Report Available at httpslawmitedupubcovid19contacttracingprivacy- principlesrelease7
Editorial Board 29 April 2020 ldquoShow Evidence That Apps for COVID-19 Contact-Tracing Are Secure and Effectiverdquo Nature 580 (7805) 563ndash563 httpsdoiorg101038d41586-020-01264-1
ldquoJoint Statement on Contact Tracingrdquo 19 April 2020 Available at httpscryptobriefingcomwp-contentuploads202004Joint-State-ment-from-Researcherspdf
Hinch Robert Will Probert Anel Nurtay Michelle Kendall Chris Wymant
108 Resources
Matthew Hall Katrina Lythgoe et al 16 April 2020 ldquoEffective Configura-tions of a Digital Contact Tracing App A Report to NHSXrdquo Available at httpsgithubcomBDI-pathogenscovid-19_instant_tracingblobmasterReportndashEffective Configurations of a Digital Contact Tracing Apppdf
Kahn Gilmor Daniel 16 April 2020 ldquoPrinciples for Technology-Assisted Contact-Tracingrdquo White Paper American Civil Liberties Union httpswwwacluorgreportaclu-white-paper-principles-technology-assisted- contact-tracing
Carroll Anna and Samantha Stroman 16 April 2020 ldquoFind My Friends in a Pandemic The Future of Contact Tracing in Americardquo CSIS Commis-sion on Strengthening Americarsquos Health Security April 16 2020 httpshealthsecuritycsisorgarticlesfind-my-friends-in-a-pandemic-the-future- of-contact-tracing-in-america
Electronic Privacy Information Center Testimony to Congress 15 April 2020 ldquoEPIC to Congress Establish Privacy Safeguards for Digital Contact Tracingrdquo httpsepicorgtestimonycongressEPIC-HEC-Contact-Tracing-Apr2020pdf
Crocker Andrew Kurt Opsahl and Bennett Cyphers 10 April 2020 ldquoThe Chal-lenge of Proximity Apps for COVID-19 Contact Tracingrdquo Electronic Fron-tier Foundation April 10 2020 httpswwwefforgdeeplinks202004challenge-proximity-apps-covid-19-contact-tracing
Hart Vi Divya Siddarth Bethan Cantrell Lila Tretikov Peter Eckersley John Langford Scott Leibrand et al 3 April 2020 ldquoOutpacing the Virus Digital Response to Containing the Spread of COVID-19 While Mitigat-ing Privacy Risksrdquo Whitepaper 5 COVID-19 Rapid Response Impact Initiative Edmond J Safra Center for Ethics httpsdrivegooglecomfiled1vIN2AX-DDNW-S0aHq8xs0RJ2jkR_CckXview
Raskar Ramesh Isabel Schunemann Rachel Barbar Kristen Vilcans Jim Gray Praneeth Vepakomma Suraj Kapa Andrea Nuzzo Rajiv Gupta et al 19 March 2020 ldquoApps Gone Rogue Maintaining Personal Privacy in an Epidemicrdquo White Paper Private Kit MIT httpsarxivorgpdf2003 08567pdf
General (not COVID-19 specific) Statements of Principles Legal Frameworks Other Information
Nuffield Council on Bioethics 2020 ldquoGuide to the Ethics of Surveillance and Quarantine for Novel Coronavirusrdquo httpswwwnuffieldbioethicsorgassetspdfsGuide-to-the-ethics-of-surveillance-and-quarantine-for-novel-coronaviruspdf
Resources 109
Schwartz Adam 2020 ldquoHow EFF Evaluates Government Demands for New Surveillance Powersrdquo Electronic Frontier Foundation April 3 2020 httpswwwefforgdeeplinks202004how-eff-evaluates-government-demands- new-surveillance-powers
WHO 2 April 2020 ldquoCoronavirus Disease 2019 (COVID-19)rdquo Situation Report 73 World Health Organization httpswwwwhointdocsdefault-sourcecoronavirusesituation-reports20200402-sitrep-73-covid-19pdf
Nuffield Council on Bioethics 17 March 2020 ldquoEthical Considerations in Responding to the COVID-19 Pandemicrdquo Rapid Policy Briefing Nuffield Council on Bioethics httpswwwnuffieldbioethicsorgassetspdfs Ethical-considerations-in-responding-to-the-COVID-19-pandemicpdf
Nuffield Council on Bioethics 28 January 2020 ldquoResearch in Global Health Emergencies Ethical Issuesrdquo httpswwwnuffieldbioethicsorgassetspdfsRGHE_full_report1pdf
Vota Wayan 4 December 2019 ldquoEvery African Countryrsquos National eHealth Strategy or Digital Health Policyrdquo ICT Works httpswwwictworksorgafrican-national-ehealth-strategy-policy
ENISA 3 December 2019 ldquoPseudonymisation Techniques and Best Practicesrdquo European Union Agency for Cybersecurity httpswwwenisaeuropaeupublicationspseudonymisation-techniques-and-best-practices
ldquoMobile Cellular Subscriptions (per 100 People)rdquo 2018 World Telecommuni-cationICT Development Report International Telecommunication Union httpsdataworldbankorgindicatorITCELSETSP2
Levinson-Waldman Rachel 2018 ldquoCellphones Law Enforcement and the Right to Privacyrdquo Brennan Center for Justice New York University School of Law httpswwwbrennancenterorgsitesdefaultfiles2019-08Report_Cell_Surveillance_Privacypdf
Code of Federal Regulations Part 46ndashProtection of Human Subjects (45 CFR 46 US Department of Health and Human Services)
General Data Protection Regulation (REGULATION (EU) 2016679 (GDPR) European Union)
European Data Protection Supervisor ldquoNecessity and Proportionalityrdquo httpsedpseuropaeudata-protectionour-worksubjectsnecessity- proportionality_en
ISOIEC 38505-12017 Information technologymdashGovernance of ITmdashGover-nance of datamdashPart 1 Application of ISOIEC 38500 to the governance of data (2017 International Organization for Standardization (ISO))
WHO 2016 ldquoWHO Guidance for Managing Ethical Issues in Infectious Dis-ease Outbreaksrdquo World Health Organization httpswwwwhointethicspublicationsinfectious-disease-outbreaksen
110 Resources
CIOMS 2016 ldquoInternational Ethical Guidelines for Health-Related Research Involving HumansndashGuideline 22 Use of Data Obtained from the Online Environment and Digital Tools in Health Related Researchrdquo Council for International Organizations of Medical Sciences httpsciomschwp- contentuploads201701WEB-CIOMS-EthicalGuidelinespdf
Principles for Digital Development 2016 ldquoPrinciplesrdquo httpsdigitalprinciplesorgprinciples
GA4GH 2014 ldquoFramework for Responsible Sharing of Genomic and Health-Related Datardquo Global Alliance for Genomics amp Health 9 Decem-ber 2014 httpswwwga4ghorggenomic-data-toolkitregulatory-ethics- toolkitframework-for-responsible-sharing-of-genomic-and-health-related- data
WHO and PATH 2013 ldquoPlanning an Information Systems Project A Toolkit for Public Health Managersrdquo World Health Organization amp PATH httpspathazureedgenetmediadocumentsTS_opt_ict_toolkitpdf
ASTHO 2012 ldquoPublic Health Collection Use Sharing and Protection of Informationrdquo Issue Brief ASTHO Legal Preparedness Series Association of State and Territorial Health Officials httpswwwasthoorgProgramsPreparednessPublic-Health-Emergency-LawPublic-Health-and-Informat- ion-Sharing-ToolkitCollection-Use-Sharing-and-Protection-Issue-Brief
Cavoukian Ann 2010 ldquoPrivacy by Design The 7 Foundational Principlesrdquo Ontario Canada Information and Privacy Commissioner of Canada httpsiapporgmediapdfresource_centerpbd_implement_7found_ principlespdf
PHLS 2002 ldquoPrinciples of the Ethical Practice of Public Healthrdquo Public Health Leadership Society httpswwwaphaorg-mediafilespdfmembergroupsethicsethics_brochureashx
National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research 1979 ldquoThe Belmont Report Ethical Principles and Guidelines for the Protection of Human Subjects of Researchrdquo US Dept of Health amp Human Services Office of Human Research Protections httpswwwhhsgovohrpregulations-and-policybelmont-reportread-the-bel mont-reportindexhtml
Reopening ProposalsPlans (Broad)
United Kingdom Cabinet Office 2020 ldquoOur Plan to Rebuild The UK Govern-mentrsquos COVID-19 Recovery Strategyrdquo 12 May 2020 httpswwwgovukgovernmentpublicationsour-plan-to-rebuild-the-uk-governments-covid- 19-recovery-strategyour-plan-to-rebuild-the-uk-governments-covid-19- recovery-strategyfourteen-supporting-programmes
Resources 111
Shannon Joel Lorenzo Reyes and Doyle Rice 2020 ldquoAre Lockdowns Being Relaxed in My State Herersquos How America Is Reopening amid the Coronavirus Pandemicrdquo USA TODAY May 21 2020 httpswww usatodaycomstorynewshealth20200419coronavirus-lockdown- reopening-states-us-texas-florida5155269002
Romer Paul 2020 ldquoRoadmap to Responsibly Reopen Americardquo 23 April 2020 httpsroadmappaulromernetpaulromer-roadmap-reportpdf
Governor Larry Hogan 24 April 2020 ldquoMaryland Strong Roadmap to Recoveryrdquo httpsgovernormarylandgovwp-contentuploads202004MD_Strongpdf
NGA and ASTHO 21 April 2020 ldquoRoadmap to Recovery A Public Health Guide for Governorsrdquo National Governors Association and American State and Territorial Health Officials httpswwwngaorgwp-contentuploads202004NGA-Reportpdf
Allen Danielle Julius Krein Ganesh Sitaraman and E Glen Weyl 2020 ldquoNational Covid-19 Testing Action Planrdquo The Rockefeller Foundation httpswwwrockefellerfoundationorgwp-contentuploads202004TheRockefellerFoundation_WhitePaper_Covid19_4_22_2020pdf
Edmond J Safra Center for Ethics Harvard UniverityndashWhite PapersAllen Danielle Sharon Block Joshua Cohen et al 20 April 2020 ldquoRoad-
map to Pandemic Resiliencerdquo httpsethicsharvardedufilescenter- for-ethicsfilesroadmaptopandemicresilience_updated_42020_0pdf
Allen Danielle Lucas Stanczyk Rajiv Sethi Glen Weyl 25 March 2020 ldquoWhen Can We Go Outrdquo httpsdrivegooglecomfiled1gf21eYeNWwrR9OO5nzxn1jlv-RTmHkt0view
Mulheirn Ian Sam Alvis Lizzie Insall James Browne Christina Palmou 20 April 2020 ldquoA Sustainable Exit Strategy Managing Uncertainty Minimis-ing Harmrdquo Tony Blair Institute for Global Change httpsinstituteglobalsitesdefaultfilesinline-filesA20Sustainable20Exit20Strategy2C 20Managing20Uncertainty20Minimising20Harmpdf
Emanuel Zeke Neera Tanden Adam Conner Erin Simpson Nicole Rap-fogel and Maura Calsyn 2020 ldquoA National and State Plan to End the Coronavirus Crisisrdquo Center for American Progress April 3 2020 httpswwwamericanprogressorgissueshealthcarenews20200403482613national-state-plan-end-coronavirus-crisis
Gottlieb Scott Caitlin Rivers Mark McClellan Lauren Silvis and Crystal Watson 2020 ldquoNational Coronavirus Response A Road Map to Reopen-ingrdquo American Enterprise Institute httpswwwaeiorgresearch-productsreportnational-coronavirus-response-a-road-map-to-reopening
112 Resources
Digital Contact Tracing Experiences from Other Countries
Multiple
Woodhams Samuel 12 May 2020 ldquoCOVID-19 Digital Rights Trackerrdquo Top10VPN httpswwwtop10vpncomresearchinvestigationscovid-19- digital-rights-tracker
Fahim Kareem Min Joo Kim and Steve Hendrix 2 May 2020 ldquoCellphone Monitoring Is Spreading with the Coronavirus So Is an Uneasy Tolerance of Surveillancerdquo Washington Post (Washington DC) httpswww washingtonpostcomworldcellphone-monitoring-is-spreading-with-the- coronavirus-so-is-an-uneasy-tolerance-of-surveillance20200502 56f14466-7b55-11ea-a311-adb1344719a9_storyhtml
Ikram Umar Christer Mjaringset MD Anne-Marie Boxall Mylaine Breton Ines Gravey Holly Krelle Veacuteronique Raimond and Reginald D Williams II 30 April 2020 ldquoWhat Can the US Learn from Innovative Strategies Used in Other Countries to Respond to COVID-19rdquo The Commonwealth Fund httpswwwcommonwealthfundorgblog2020what-can-us-learn-innovative-strategies-used-other-countries-respond-covid-19
Jens-Henrik Jeppesen and Pasquale Esposito 29 April 2020 ldquoCOVID-19 European Data Collection and Contact Tracing Measuresrdquo Center for Democracy amp Technology httpscdtorginsightscovid-19-european- data-collection-and-contact-tracing-measures
Africa CDC 9 April 2020 ldquoGuidance on Contact Tracing for COVID-19 Pan-demicrdquo Manuals Guidelines amp Frameworks African Union Africa CDC httpsafricacdcorgdownloadguidance-on-contact-tracing-for-covid-19- pandemic
Heneghan Carl Jon Brassey and Tom Jefferson 6 April 2020 ldquoCOVID-19 What Proportion Are Asymptomaticrdquo Centre for Evidence-Based Medi-cine httpswwwcebmnetcovid-19covid-19-what-proportion-are- asymptomatic
Kharpal Arjun 30 March 2020 ldquoUse of Surveillance to Fight Coronavirus Raises Concerns about Government Power after Pandemic Endsrdquo CNBC httpswwwcnbccom20200327coronavirus-surveillance-used-by- governments-to-fight-pandemic-privacy-concernshtml
Canada
ldquoCommissioner Publishes Framework to Assess Privacy-Impactful Initiatives in Response to COVID19rdquo Office of the Privacy Commissioner of Canada 17 April 2020 httpswwwprivgccaenopc-newsnews-and-announcements 2020an_200417
Resources 113
China
Kraemer Moritz U G Chia-Hung Yang Bernardo Gutierrez Chieh-Hsi Wu Brennan Klein David M Pigott Open COVID-19 Data Working Group Louis du Plessis Nuno R Faria Ruoran Li William P Hanage John S Brownstein Maylis Layan Alessandro Vespignani Huaiyu Tian Chris-topher Dye Oliver G Pybus Samuel V Scarpino ldquoThe Effect of Human Mobility and Control Measures on the COVID-19 Epidemic in Chinardquo Science 368(6490) 493-497 DOI 101126scienceabb4218
Bi Qifang Yongsheng Wu Shujiang Mei Chenfei Ye Xuan Zou Zhen Zhang Xiaojian Liu Lan Wei Shaun A Truelove Tong Zhang Wei Gao Cong Cheng Xiujuan Tang Xiaoliang Wu Yu Wu Binbin Sun Suli Huang Yu Sun Juncen Zhang Ting Ma Justin Lessler and Teijian Feng ldquoEpidemiol-ogy and Transmission of COVID-19 in 391 Cases and 1286 of Their Close Contacts in Shenzhen China A Retrospective Cohort Studyrdquo The Lancet April 27 2020 DOIhttpsdoiorg101016S1473-3099(20)30287-5
Sun Kaiyuan and Ceacutecile Viboud ldquoImpact of contact tracing on SARS-CoV-2 transmissionrdquo The Lancet April 27 2020 DOIhttpsdoiorg101016S1473-3099(20)30357-1
Ferretti Luca Chris Wymant Michelle Kendall Lele Zhao Anel Nurtay Lucie Abeler-Doumlrner Michael Parker David Bonsall and Christophe Fraser 2020 ldquoQuantifying SARS-CoV-2 Transmission Suggests Epidemic Control with Digital Contact Tracingrdquo Science 368 (6491) httpsdoiorg101126scienceabb6936
Gan Nectar and David Culver 2020 ldquoChina Is Fighting the Coronavirus with a Digital QR code Herersquos How It Worksrdquo CNN Business April 16 2020 httpswwwcnncom20200415asiachina-coronavirus-qr-code-intl-hnkindexhtml
Mozur Paul Raymond Zhong and Aaron Krolik 2020 ldquoIn Coronavirus Fight China Gives Citizens a Color Code with Red Flagsrdquo New York Times (New York NY) March 1 2020 httpswwwnytimescom20200301businesschina-coronavirus-surveillancehtml
Germany
Schwartz Matthew S 2020 ldquoGermany Backs Away from Compiling Coro-navirus Contacts in a Central Databaserdquo NPR April 27 2020 https wwwnprorgsectionscoronavirus-live-updates20200427846046185 germany-backs-away-from-compiling-coronavirus-contacts-in-a-central- databaseutm_medium=RSSamputm_campaign=news
Busvine Douglas and Andreas Rinke 2020 ldquoGermany Flips to Apple-Google
114 Resources
Approach on Smartphone Contact Tracingrdquo Reuters April 26 2020 httpswwwreuterscomarticleus-health-coronavirus-europe-techgermany-flips-on-smartphone-contact-tracing-backs-apple-and-google-idUSKCN22807J
Hong Kong
ldquo lsquoStayHomeSafersquo Mobile App User Guiderdquo The Government of the Hong Kong
Special Administrative Region May 20 2020httpswwwcoronavirusgov hkengstay-home-safehtml
India
Dixit Pranav 2020 ldquoIndiarsquos Contact Tracing App Is All But Mandatory So This Programmer Hacked It So That He Always Appears Saferdquo BuzzFeed News May 12 2020 httpswwwbuzzfeednewscomarticlepranavdixitindia-aarogya-setu-hacked
Greenberg Andy 2020 ldquoIndiarsquos Covid-19 Contact Tracing App Could Leak Patient Locationsrdquo WIRED May 6 2020 httpswwwwiredcomstoryindia-covid-19-contract-tracing-app-patient-location-privacy
Alderson Elliot ldquoAarogya Setu The Story of a Failurerdquo Medium May 6 2020 httpsmediumcomfs0c131yaarogya-setu-the-story-of-a-failure- 3a190a18e34
OrsquoNeill Patrick Howell ldquoIndia Is Forcing People to Use Its COVID App Unlike Any Other Democracyrdquo MIT Technology Review May 6 2020 httpswwwtechnologyreviewcom202005071001360india-aarogya- setu-covid-app-mandatory
Israel
Hendrix Steve and Ruth Eglash 2020 ldquoIsrael Is Using Cellphone Surveil-lance to Warn Citizens You May Already Be Infectedrdquo Washington Post (Washington DC) March 19 2020 httpswwwwashingtonpostcomworldmiddle_eastisrael-is-using-cellphone-surveillance-to-warn-citizens-you-may-already-be-infected2020031968267294-69e7-11ea-b199-3a9799c54512_storyhtml
Singapore
ldquoTraceTogether Safer Togetherrdquo A Singapore Government Agency Website accessed May 21 2020 httpswwwtracetogethergovsg
Resources 115
South Korea
Korean Ministry of Health and Welfare ldquoConfirmed Patient Movement Path Websiterdquo Central Accident Remediation Headquarters accessed May 21 2020 httpncovmohwgokrbdBoardList_RealdobrdId=1ampbrdGu-bun=12ampncvContSeq=ampcontSeq=ampboard_id=ampgubun=
Kim Max S 2020 ldquoSeoulrsquos Radical Experiment in Digital Contact Tracingrdquo The New Yorker April 17 2020 httpswwwnewyorkercomnewsnews-deskseouls-radical-experiment-in-digital-contact-tracing
Watson Ivan and Sophie Jeong 2020 ldquoCoronavirus Mobile Apps Are Surging in Popularity in South Koreardquo CNN Business February 28 2020 httpseditioncnncom20200228techkorea-coronavirus-tracking-appsindexhtml
United Kingdom
Hern Alex and Kate Proctor 2020 ldquoUK May Ditch NHS Contact-Tracing App for Apple and Google Modelrdquo The Guardian May 7 2020 httpswwwtheguardiancomtechnology2020may07uk-may-ditch-nhs-contact- tracing-app-for-apple-and-google-model
Lovejoy Ben 2020 ldquoHands-on with UKrsquos NHS Contact Tracing App as the Test Goes Liverdquo 9-5 Mac Blog May 7 2020 https9to5maccom2020 0507nhs-contact-tracing
ldquoCoronavirus Test Track and Trace Plan Launched on Isle of Wightrdquo Depart-ment of Health and Social Care Press Release May 4 2020 httpswww govukgovernmentnewscoronavirus-test-track-and-trace-plan-launched- on-isle-of-wight
Leprince-Ringuet Daphne 2020 ldquoContact-Tracing Apps Why the NHS Said No to Apple and Googlersquos Planrdquo ZDNet April 28 2020 httpswww zdnetcomarticlecontact-tracing-apps-why-the-nhs-said-no-to-apple- and-googles-plan
Specific Digital ProductsApps
COVID-19
Vota Wayan 2020 Additional Proposed Coronavirus Solutions Google Doc-ument httpsdocsgooglecomspreadsheetsd15hkhdtGNzx7oHkO8Y 2MOiY83JsHjqxL4MhMGvlA_J6Ieditgid=579623365
Apple and Google nd ldquoPrivacy-Preserving Contact Tracingrdquo Apple Accessed May 19 2020 httpswwwapplecomcovid19contacttracing
116 Resources
Woodhams Samuel 12 May 2020 ldquoCOVID-19 Digital Rights Trackerrdquo Top10VPN httpswwwtop10vpncomresearchinvestigationscovid-19-digital-rights-tracker
Starobinski David and Johannes Becker 30 April 2020 ldquoHow Apple and Goo-gle Will Let Your Phone Warn You If Yoursquove Been Exposed to the Corona-virusrdquo The Conversation Accessed May 21 2020 httptheconversation comhow-apple-and-google-will-let-your-phone-warn-you-if-youve-been-exposed-to-the-coronavirus-136597
Morrison Sara 2020 ldquoThe Apple-Google Contact Tracing Tool Gets a Beta Release and a New Risk Level Featurerdquo Vox April 24 2020 httpswwwvoxcomrecode202042421234420apple-google-contact-tracing- exposure-notification-update
Commonwealth Centre for Digital Health 9 April 2020 ldquo[Webinar] CWCDH Digital Response to COVID-19rdquo 5040 httpswwwyoutubecomwatchv=ZyE_KRWLtC8ampfeature=youtube
Johns Hopkins Medicine 23 April 2020 ldquoJohns Hopkins Medicine Remote Monitoring Program for Faculty Staff and Providers Exposed to COVID-19rdquo Johns Hopkins Medicine Occupational Health Services httpswww hopkinsmedicineorghsecovid19_emocha
MIT nd ldquoPrivate Kit Safe Paths Privacy-by-Design Covid19 Solutions Using GPS+Bluetooth for Citizens and Public Health Officialsrdquo Safepaths Accessed May 20 2020 httpssafepathsmitedu
Related Disease Detection Projects
ldquoCOVID Controlrdquo Google Play accessed May 21 2020 httpsplaygoogle comstoreappsdetailsid=jhueduJohnsHopkinsCOVIDControl
ldquoCOVID Symptom Trackerrdquo Created by Massachusetts General Hospital the Harvard TH Chan School of Public Health Kingrsquos College London and Stanford University School of Medicine accessed May 21 2020 httpscovidjoinzoecomus
ldquoSee How Your Community Is Moving around Differently Due to COVID-19rdquo Google COVID-19 Mobility Reports accessed May 21 2020 httpswww googlecomcovid19mobility
Drew David A Long H Nguyen Claire J Steves Cristina Menni Maxim Freydin Thomas Varsavsky Carole H Sudre M Jorge Cardoso Sebastien Ourselin Jonathan Wolf Tim D Spector Andrew T Chan and COPE Consortium 2020 ldquoRapid Implementation of Mobile Technology for Real-Time Epidemiology of COVID-19rdquo Science 05 May 2020 DOI 101126scienceabc0473
Resources 117
Tress Luke 2020 ldquoMaccabi Medial EarlySign Develop Algorithm to Identify High-Risk COVID-19 Casesrdquo Times of Israel April 22 2020 httpswwwtimesofisraelcommaccabi-medial-earlysign-develop-algo rithm-to-identify-high-risk-covid-19-cases
ldquoPCR Diagnostic Testing for SARS-CoV-2rdquo Center for Health Security Johns Hopkins Bloomberg School of Public Health last modified April 17 2020 httpswwwcenterforhealthsecurityorgresourcesCOVID-19COVID-19-fact-sheets200130-nCoV-diagnostics-factsheetpdf
Radin Jennifer M Nathan E Wineinger Eric J Topol and Steven R Steinhubl 2020 ldquoHarnessing Wearable Device Data to Improve State-Level Real-Time Surveillance of Influenza-Like Illness in the USA A Population-Based Studyrdquo The Lancet January 16 2020 DOIhttpsdoiorg101016S2589-7500(19)30222-5
Polling
Hargittai Eszter Minh Hao Nguyen Jaelle Fuchs Jonathan Gruber Will Marler Amanda Hunsaker and Goumlkccedile Karaoglu 2020 ldquoCovid-19 Study on Digital Media and the Coronavirus Pandemicrdquo Internet Use and Society Division Institute of Communication and Media Research University of Zurich httpwebuseorgcovid
Navigator Research 2020 ldquoPublic Opinion on Coronavirus Navigator Updaterdquo Navigating Coronavirus (blog) May 21 2020 https navigatorresearchorgnavigating-coronavirus
Russonello Giovanni 2020 ldquoBig Government For Now Most Americans Say Bring It Onrdquo The New York Times May 1 2020 sec US httpswww nytimescom20200501uspoliticscoronavirus-spending-pollshtml
ldquoWashington Post-University of Maryland National Pollrdquo 2020 Washington Post April 21 2020 httpswwwwashingtonpostcomcontext washington-post-university-of-maryland-national-poll-april-21-26-20203583b4e9-66be-4ed6-a457-f6630a550ddf
Kirzinger Ashley Liz Hamel Cailey Muntildeana Audrey Kearney and Mollyann Brodie 2020 ldquoKFF Health Tracking Poll ndash Late April 2020 Coronavirus Social Distancing and Contact Tracingrdquo Kaiser Family Foundation (blog) April 24 2020 httpswwwkfforgcoronavirus-covid-19issue-briefkff-health-tracking-poll-late-april-2020
NYC DOH 2020 ldquoAge-Adjusted Rates of Lab Confirmed COVID-19 Non-hospitalized Cases Estimated Non-Fatal Hospitalized Cases and Persons Known to Have Died per 100000 by RaceEthnicity Grouprdquo New York
118 Resources
City Department of Health httpswww1nycgovassetsdohdownloadspdfimmcovid-19-deaths-race-ethnicity-04242020-1pdf
Anderson Monica and Brooke Auxier 2020 ldquoMost Americans Donrsquot Think Cellphone Tracking Will Help Limit COVID-19 Are Divided on Whether Itrsquos Acceptablerdquo Pew Research Center (blog) April 16 2020 httpswwwpewresearchorgfact-tank20200416most-americans-dont-think-cellphone-tracking-will-help-limit-covid-19-are-divided-on-whether-its-acceptable
Daly Kyle 2020 ldquoExclusive Americans Wary of Giving up Data to Fight Coronavirusrdquo Axios April 3 2020 httpswwwaxioscomexclusive- americans-wary-of-giving-up-data-to-fight-coronavirus-330fc1d9-8b99-4a51-871b-25ee0e0591f2html
Elliott Douglas J Ana Kreacic Lorenzo Milans del Bosch and Lisa Quest nd ldquoData-Sharing in the Time of Coronavirusrdquo Oliver Wyman Forum Accessed May 19 2020 httpswwwoliverwymanforumcomfuture-of-data2020aprdata-sharing-in-the-time-of-coronavirushtml
Auxier Brooke Lee Rainie Monica Anderson Andrew Perrin Madhu Kumar and Erica Turner 2019 ldquoAmericans and Privacy Concerned Confused and Feeling Lack of Control Over Their Personal Informationrdquo Pew Research
Center Internet Science amp Tech (blog) November 15 2019 httpswww pewresearchorginternet20191115americans-and-privacy-concerned- confused-and-feeling-lack-of-control-over-their-personal-information
EPIC 2020 ldquoPublic Opinion on Privacyrdquo Electronic Privacy Information Center January 22 2020 httpsepicorgprivacysurvey
Pew Research Center 2020 ldquoDemographics of Mobile Device Ownership and Adoption in the United Statesrdquo Internet amp Technology Pew Research Cen-ter httpswwwpewresearchorginternetfact-sheetmobile
Rodrigues Rafaela Alina Husain Amanda Couture-Carron Leslye E Orloff and Nawal H Ammar 2018 ldquoPromoting Access to Justice for Immigrant and Limited English Proficient Crime Victims in an Age of Increased Immi-gration Enforcement Initial Report from a 2017 National Surveyrdquo Wash-ington DC National Immigrant Womenrsquos Advocacy Project American University Washington College of Law httpniwaplibrarywclamericaneduwp-contentuploadsImmigrant-Access-to-Justice-National-Reportpdf
Center for Survey Measurement (CSM) MEMORANDUM FOR Associate Directorate for Research and Methodology (ADRM) 2017 ldquoRespondent Confidentiality Concernsrdquo September 20 2017 httpswww2censusgovcacnacmeetings2017-11Memo-Regarding-Respondent-Confidentiality- Concernspdf
Resources 119
Pew Research Center 2017 ldquoUS Muslims Concerned About Their Place in Society but Continue to Believe in the American Dreamrdquo Religion amp Public Life Pew Research Center httpswwwpewforumorg20170726findings-from-pew-research-centers-2017-survey-of-us-muslims
Popular Press
Dixit Pranav 2020 ldquoIndiarsquos Contact Tracing App Is All But Mandatory So This Programmer Hacked It So That He Always Appears Saferdquo BuzzFeed News May 12 2020
Mills Rodrigo Chris 2020 ldquoDigital Contact Tracing Is Becoming Available but Is It Effectiverdquo The Hill May 7 2020 httpsthehillcompolicytechnology 496498-digital-contact-tracing-is-becoming-available-but-is-it-effective
Ingram David 2020 ldquoApple Google Push Makers of Coronavirus Apps Not to Record User Locationrdquo May 4 2020 NBC News Accessed May 19 2020 httpswwwnbcnewscomtechtech-newscoronavirus-apps-won-t-be-able- record-users-location-apple-n1199586
McMinn Sean 2020 ldquoMobile Phone Data Show More Americans Are Leav- ing Their Homes Despite Ordersrdquo NPR May 1 2020 httpswwwnpr org20200501849161820mobile-phone-data-show-more-americans- are-leaving-their-homes-despite-orders
Valentino-DeVries Jennifer Natasha Singer and Aaron Krolik 2020 ldquoA Scram-ble for Virus Apps That Do No Harmrdquo The New York Times April 29 2020 sec Business httpswwwnytimescom20200429 businesscoronavirus-cellphone-apps-contact-tracinghtml
Giglio Mike 2020 ldquoWould You Sacrifice Your Privacy to Get Out of Quaran-tinerdquo The Atlantic April 22 2020 httpswwwtheatlanticcompoliticsarchive202004coronavirus-pandemic-privacy-civil-liberties-911609172
Bradshaw Tim 2020 ldquo2 Billion Phones Cannot Use Google and Apple Con-tact-Tracing Tech | Ars Technicardquo Ars Technica April 20 2020 httpsarstechnicacomtech-policy2020042-billion-phones-cannot-use-google-and-apple-contract-tracing-tech
Reston Maeve Kristina Sgueglia and Cheri Mossburg 2020 ldquoGovernors on East and West Coasts Form Pacts to Decide When to Reopen Economiesrdquo CNN Politics April 13 2020 httpswwwcnncom20200413politicsstates-band-together-reopening-plansindexhtml
Cochrane Emily Claire Cain Miller and Jim Tankersley 2020 ldquoTrump Admin-istration Scales Back Paid Leave in Coronavirus Relief Lawrdquo The New York Times April 2 2020 sec US httpswwwnytimescom20200402uspoliticscoronavirus-paid-leavehtml
120 Resources
Kim Nemo 2020 ldquolsquoMore Scary than Coronavirusrsquo South Korearsquos Health Alerts Expose Private Livesrdquo The Guardian March 6 2020 sec World news httpswwwtheguardiancomworld2020mar06more-scary- than-coronavirus-south-koreas-health-alerts-expose-private-lives
Hickey Matt 2014 ldquoCarriers Can Now Install Apps On Android Handsets Without Customersrsquo Permissionrdquo Forbes December 1 2014 httpswww forbescomsitesmatthickey20141201carriers-can-now-install-apps-on- android-handsets-without-customers-permission
Commentaries
Giubilini Alberto 2020 ldquoContact-Tracing Apps and the Future COVID-19 Vaccination Should Be Compulsory Social Technological and Pharmaco-logical Immunisationrdquo Practical Ethics (blog) May 6 2020 httpblo gpracticalethicsoxacuk202005contact-tracing-apps-and-the-future- covid-19-vaccination-should-be-compulsory-social-technological-and- pharmacological-immunisation
Landau Susan Christy Lopez and Laura Moy 2020 ldquoThe Importance of Equity in Contact Tracingrdquo Lawfare (blog) May 1 2020 httpswww lawfareblogcomimportance-equity-contact-tracing
Schaefer G Owen and Angela Ballantyne 2020 ldquoDownloading COVID-19 Contact Tracing Apps Is a Moral Obligationrdquo Journal of Medical Ethics
Blog (blog) May 4 2020 httpsblogsbmjcommedical-ethics20200504downloading-covid-19-contact-tracing-apps-is-a-moral-obligation
OrsquoNeill Patrick Howell 2020 ldquoFive Things We Need to Do to Make Contact Tracing Really Workrdquo MIT Technology Review April 28 2020 httpswwwtechnologyreviewcom202004281000714five-things-to-make- contact-tracing-work-covid-pandemic-apple-google
Doffman Zak 2020 ldquoCOVID-19 Contact Tracing Why Apple And Google Canrsquot Make This Workrdquo Forbes April 27 2020 httpswwwforbescomsiteszakdoffman20200427this-is-the-contact-tracing-worry-even-apple- and-google-cant-resolve
All Tech is Human 2020 The Ethics of Contact Tracing for COVID-19 httpswwwyoutubecomwatchv=59mKUAVDhdkampt=626s
Canca Cansu 2020 ldquoWhy lsquoMandatory Privacy-Preserving Digital Contact Tracingrsquo Is the Ethical Measure against COVID-19rdquo Medium May 4 2020 httpsmediumcomcansucancawhy-mandatory-privacy-preserv-ing-digital-contact-tracing-is-the-ethical-measure-against-covid-19-a0d-143b7c3b6
Goodman Bryce 2020 ldquoCOVID and Contact Tracing When Social Justice
Resources 121
Demands Mass Surveillancerdquo Medium April 10 2020 httpsmediumcombwgoodmancovid-and-contact-tracing-when-social-justice-demands-mass-surveillance-18d419b8cc5
Gray Rosie and Caroline Haskins 2020 ldquoThey Were Opposed To Govern-ment Surveillance Then The Coronavirus Pandemic Beganrdquo BuzzFeed News March 30 2020 httpswwwbuzzfeednewscomarticlerosiegraythey-were-opposed-to-government-surveillance-then-the
Cegłowski Maciej 2020 ldquoWe Need A Massive Surveillance Programrdquo Idle
Words (blog) March 23 2020 httpsidlewordscom202003we_need_ a_massive_surveillance_programhtm
Academic Literature
COVID-19 Specific
Abeler J Baumlcker M Buermeyer U Zillessen H (2020) COVID-19 Contact Tracing and Data Protection Can Go Together JMIR mHealth and uHealth 8(4) e19359 doi 10219619359
Altmann S Milsom L Zillessen H et al (2020) Acceptability of App-Based Contact Tracing for COVID-19 Cross-Country Survey Evidence Preprint
Berke A Bakker M Vepakomma P Larson K Pentland A (2020) Assessing Disease Exposure Risk with Location Data A Proposal for Cryptographic Preservation of Privacy arXiv arXiv200314412ndashMarch 2020
Bonsail D Parker M Fraser C (2020) Sustainable Containment of COVID-19 Using Smartphones in China Scientific and Ethical Underpinnings for Implementation of Similar Approaches in Other Settings Unpublished working paper
Bradshaw WJ Alley EC Huggins JH Lloyd AL Esvelt KM (2020) Bidirectional Contact Tracing Is Required for Reliable COVID-19 Control Preprint via MedRxiv
Braithwaite I Callender T Bullock M Aldridge R (2020) Automated and Semi-Automated Contact Tracing Protocol for a Rapid Review of Avail-able Evidence and Current Challenges to Inform the Control of COVID-19 Preprint via medRxiv doi httpsdoiorg1011012020041420063636
Bulchandani Bannerjee V Shivam S Moudgalya S Sondhi SL (2020) Digital Herd Immunity and COVID-19 Preprint via medRxiv doi httpsdoiorg1011012020041520066720
Cheng H Jian S Liu D (2020) Contact Tracing Assessment of COVID-19 Transmission Dynamics in Taiwan and Risk at Different Exposure Periods Before and After Symptom Onset JAMA Internal Medicine doi101001jamainternmed20202020
122 Resources
Cho H Ippolito D Yu YW (2020) Contact Tracing Mobile Apps for COVID-19 Privacy Considerations and Related Trade-offs httpsarxivorgpdf200311511pdf
Devakumar D Geordan S Bhopal SS Abubakar I (2020) Racism and dis-crimination in COVID-19 responses The Lancet 395(10231) 1194 doi 101016S0140-6736(20)30792-3
Drew D Nguyen L Steves C et al (2020) Rapid Implementation of Mobile Technology for Real-Time Epidemiology of COVID-19 Science published online May 5 2020
Ferretti L Wymant C Kendall M et al (2020) Quantifying SARS-CoV-2 Trans-mission Suggests Epidemic Control with Digital Contact Tracing Science doi 101126scienceabb6936
Fraser C Abeler-Doumlrner L Ferretti L et al (2020) Digital Contact Tracing Comparing the Capabilities of Centralised and Decentralised Data Archi-tectures to Effectively Suppress the COVID-19 Epidemic While Maximizing Freedom of Movement and Maintaining Privacy Preprint
Leith DJ Farrell S (2020) Coronavirus Contact Tracing Evaluating The Poten-tial Of Using Bluetooth Received Signal Strength For Proximity Detection Preprint
Jayant Limaye R Sauer M Ali J et al (2020) Building Trust While Influencing Online COVID-19 Content iIn the Social Media World The Lancet Digital Health
Mahmood S Hasan K Colder Carras M Labrique A (2020) Global Prepared-ness Aagainst COVID-19 We Must Leverage the Power of Digital Health JMIR Public Health Surveill 20206(2)e18980 DOI 10219618980
Mello M Wang CJ (2020) Ethics and Governance for Digital Disease Surveil-lance Science 11 May 2020 eabb9045 DOI 101126scienceabb9045
Park S Jeehyun Choi G Ko H (2020) Information TechnologyndashBased Tracing Strategy in Response to COVID-19 in South KoreamdashPrivacy Controversies JAMA doi101001jama20206602
Parker M Fraser C Abeler-Doumlrner L Bonsall D (2020) Ethics of Instantaneous Contract Tracing Using Mobile Phone Apps in the Control of the COVID-19 Pandemic Journal of Medical Ethics Published Online May 4 2020
Ethics and Digital Disease Detection
Aiello A Renson A Civich P (2020) Social Mediandash and Internet-Based Dis-ease Surveillance for Public Health Annual Review of Public Health 41 101ndash118 doi 101146annurev-publhealth-040119-094402
Ali J DiStefano M Coates McCall I et al (2019) Ethics of Mobile Phone Sur-
Resources 123
veys to Monitor Non-Communicable Disease Risk Factors in Low- and Middle-Income Countries A Global Stakeholder Survey Global Public Health 14(8) 1167ndash1181
Ali J Labrique A Gionfriddo K et al (2017) Ethics Considerations in Global Mobile Phone-Based Surveys of Noncommunicable Diseases A Concep-tual Exploration Journal of Medical Internet Research 19(5) e110 doi 102196jmir7326
Brownstein J Freifeld C Madoff L (2009) Digital Disease DetectionndashHar-nessing the Web for Public Health Surveillance New England Journal of Medicine 360(21) 2153ndash2157 doi 101056NEJMp0900702
Danquah LO Hasham N MacFarlane M et al (2019) Use of a Mobile Application for Ebola Contact Tracing and Monitoring in Northern Sierra Leone A Proof-of-Concept Study BMC Infectious Diseases 19 810
Degeling C Carter S van Oijen A et al (2020) Community Perspectives on the Benefits and Risks of Technologically Enhanced Communicable Disease Surveillance Systems A Report on Four Community Juries BMC Medical Ethics 21 31 doi 101186s12910-020-00474-6
DeJong B Badou G Luten J et al (2019) Ethical Considerations for Movement Mapping to Identify Disease Transmission Hotspots Emerging Infectious Diseases 25(7) e181421 doi 103201eid2507181421
Denecke K (2017) An Ethical Assessment Model for Digital Disease Detec-tion Technologies Life Sciences Society and Policy 13 16 doi 101186s40504-017-0062-x
Genevieve LD Martani A Wangmo T et al (2019) Participatory Disease Surveillance Systems Ethical Framework Journal of Medical Internet Research 21(5) e12273 doi10219612273
Gilbert G Degeling C Johnson J (2017) Communicable Disease Surveillance Ethics in the Age of Big Data and New Technology Asian Bioethics Review 11 173-187 doi 101007s41649-019-00087-1
Iwaya LH Li J Fischer-Hubner S et al (2019) E-Consent for Data Privacy Consent Management for Mobile Health Technologies in Public Health Surveys and Disease Surveillance Studies in health technology and infor-matics 264 1223-1227 doi 103233SHTI190421
Kostkova P (2018) Disease Surveillance Data Sharing for Public Health The Next Ethical Frontiers Life Sciences Society and Policy 14 16 doi 101186s40504-018-0078-x
Mahmood S Hasan K Colder Carras M Labrique A (2020) Global Prepared-ness against COVID-19 We Must Leverage the Power of Digital Health JMIR Public Health Surveillance 6(2)e18980
124 Resources
Smolinski MS Crawley AW Baltrusaitis K et al (2015) Flu Near You Crowd-sourced Symptom Reporting Spanning 2 Influenza Seasons AJPH 105(10) 2124ndash2130
Wojcik O Brownstein J Chunara R Johansson M (2014) Public Health for the People Participatory Infectious Disease Surveillance in the Digital Age Emerging Themes in Epidemiology 11 7 doi 1011861742-7622-11-7
Other Works
Bernstein J Holroyd TA Atwell JE et al (2019) Rockland Countyrsquos Proposed Ban against Unvaccinated Minors Balancing Disease Control Trust and Liberty Vaccine 37(30) 3933ndash3935
Berry SM Petzold EA Dull P et al (2017) A Response Adaptive Randomiza-tion Platform Trial for Efficient Evaluation of Ebola Virus Treatments A Model for Pandemic Response Clinical Trials 13(1) 22ndash30 doi 101177 1740774515621721
Beukenhorst AL Schultz DM McBeth J (2017) Using Smartphones for Research outside Clinical Settings How Operating Systems App Develop-ers and Users Determine Geolocation Data Quality in mHealth Studies MEDINFO 2017 Precision Healthcare through Informatics
Bourne P (2015) Confronting the Ethical Challenges of Big Data in Public Health Plos Computational Biology 11(2) e1004073 doi 101371journal pcbi1004073
Doerr M Suver C Wilbanks J (2016) Developing a Transparent Par ticipant-Navigated Electronic Informed Consent for Mobile-Mediated Research (April 22 2016) Available at SSRN httpsssrncomabstract=2769129 or httpdxdoiorg102139ssrn2769129
Dredze M Paul MJ Bergsma S Tran H (2013) Carmen A Twitter Geolocation System with Applications to Public Health Expanding the Boundaries of Health Informatics Using Artificial Intelligence Papers from the AAAI 2013 Workshop
Eckhoff PA Tatem AJ (2015) Digital methods in epidemiology can transform disease controlInternational Health Volume 7 Issue 2 March 2015 Pages 77ndash78 httpsdoiorg101093inthealthihv013
Edelstein M Lee L Herten-Crabb A Heymann D Harper D (2018) Strength-ening Global Public Health Surveillance through Data and Benefit Sharing Emerging Infectious Diseases 24(7) 1324ndash1330 doi 103201eid2407151830
Faden R Beauchamp T (1986) A History and Theory of Informed Consent Oxford University Press ISBN 9780199748655
Resources 125
Fairchild A Bayer R (2004) Ethics and the Conduct of Public Health Surveil-lance Science 303(5658) 631ndash632
Flanagan M Howe DC Nissenbaum H (2008) Embodying Values in Technol-ogy Theory and Practice In Information Technology and Moral Philoso-phy van den Hoven J amp Weckert J (eds) Cambridge Cambridge Univer-sity Press 322ndash353
Fraccaro P Beukenhorst A Sperrin M et al (2019) Digital Biomarkers from Geolocation Data in Bipolar Disorder and Schizophrenia A Systematic Review Journal of the American Medical Informatics Association Volume 26 Issue 11 November 2019 Pages 1412ndash1420
Furlanello C Merler S Menegon S et al (2002) New WEBGIS Technologies for Geo-location of Epidemiological Data An Application for the Sur-veillance of the Risk of Lyme borreliosis Disease In Giornale Italiano di Aritmologia e Cardiostimolazione Special Issue Proceedings of the 15th Int Congress ldquoThe New Frontiers of Arrhythmiasrdquo v 5 n1 Mar 2002 241ndash245
Gibson DG Wosu AC Pariyo GW et al (2019) Effect of Airtime Incentives on Response and Cooperation Rates in Non-communicable Disease Interactive Voice Response Surveys Randomised Controlled Trials in Bangladesh and Uganda BMJ Global Health 4(5)
Knobel C Bowker GC (2011) Computing Ethics Values in Design Communi-cations of the acm 54(7) 26ndash28
Lee R Cui RR Muessig KE et al (2015) Incentivizing HIVSTI Testing A Sys-tematic Review of the Literature AIDS and Behavior 18(5) 905ndash912 doi 101007s10461-013-0588-8
Lorchan LT Wyatt J (2014) mHealth and Mobile Medical Apps A Frame-work to Assess Risk and Promote Safer Use Journal of Medical Internet Research 16(9) e210 doi 102196jmir3133
Mathews S McShea M Hanley C et al (2019) Digital Health A Path to Vali-dation Digital Medicine 238 doi 101038s41746-019-0111-3
Moore S Tasse A Thorogood A et al (2017) Consent Processes for Mobile App Mediated Research Systematic Review JMIR mHealth amp uHealth 5(8) e126 DOI 102196mhealth7014
Pallman P Bedding AW Choodari-Oskooei B et al (2018) Adaptive Designs in Clinical Trials Why Use Them and How to Run and Report Them BMC Medicine 16(29)
Rennie S Buchbinder M Juengst E et al (2020) Scraping the Web for Public Health Gains Ethical Considerations from a lsquoBig Datarsquo Research Project on HIV and Incarceration Public Health Ethics doi 101093phephaa006
126 Resources
Rithalia A McDaid C Suekarran S (2009) Impact of Presumed Consent for Organ Donation on Donation Rates A Systematic Review BMJ 2009 338
Singer E Ye C (2012) The Use and Effects of Incentives in Surveys The ANNALS of the American Academy of Political and Social Science 645(1) 112ndash141
Vayena E Blassime A (2018) Health Research with Big Data Time for Systemic Oversight J Law Med Ethics 2018 Mar 46(1) 119ndash129
Vayena E Mastroianni A Kahn J (2012) Ethical Issues in Health Research with Novel Online Sources American Journal of Public Health 102(12) 2225ndash2230 doi 102105AJPH2012300813
Vayena E Salatheacute M Madoff L Brownstein J (2015) Ethical Challenges of Big Data in Public Health Plos Computational Biology 11(2) e1003904 doi 101371journalpcbi1003904
Xafis V Schaefer GO Labude MK et al (2019) An Ethics Framework for Big Data in Health and Research Asian Bioethics Review volume 11 pages 227ndash254(2019)
Works Cited
127
Africa CDC 2020 ldquoGuidance on Contact Tracing for COVID-19 Pandemicrdquo Manuals Guidelines amp Frameworks African Union Africa CDC httpsafricacdcorgdownloadguidance-on-contact-tracing-for-covid-19- pandemic
Aiello Allison E Audrey Renson and Paul N Zivich 2020 ldquoSocial Mediandash and Internet-Based Disease Surveillance for Public Healthrdquo Annual Review of Public Health 41 (1) 101ndash18 httpsdoiorg101146annurev-publhealth-040119-094402
Ali Joseph Michael J DiStefano Iris Coates McCall Dustin G Gibson Gulam Muhammed Al Kibria George W Pariyo Alain B Labrique and Adnan A Hyder 2019 ldquoEthics of Mobile Phone Surveys to Monitor Non-Communicable Disease Risk Factors in Low- and Middle-Income Countries A Global Stakeholder Surveyrdquo Global Public Health 14 (8) 1167ndash81 httpsdoiorg1010801744169220191566482
Ali Joseph Alain B Labrique Kara Gionfriddo George Pariyo Dustin G Gibson Bridget Pratt Molly Deutsch-Feldman and Adnan A Hyder 2017 ldquoEthics Considerations in Global Mobile Phone-Based Surveys of Noncommunicable Diseases A Conceptual Explorationrdquo Journal of Medi-cal Internet Research 19 (5) e110 httpsdoiorg102196jmir7326
Allen Danielle Julius Krein Ganesh Sitaraman and E Glen Weyl 2020 ldquoNational Covid-19 Testing Action Planrdquo The Rockefeller Foundation httpswwwrockefellerfoundationorgwp-contentuploads202004TheRockefellerFoundation_WhitePaper_Covid19_4_22_2020pdf
Altmann Samuel Luke Milsom Hannah Zillessen Raffaele Blasone Frederic Gerdon Ruben Bach Frauke Kreuter Daniele Nosenzo Severine Tous-saert and Johannes Abeler 2020 ldquoAcceptability of App-Based Contact Tracing for COVID-19 Cross-Country Survey Evidencerdquo MedRxiv May 2020050520091587 httpsdoiorg1011012020050520091587
Anderson Monica and Brooke Auxier 2020 ldquoMost Americans Donrsquot Think Cellphone Tracking Will Help Limit COVID-19 Are Divided on Whether
128 Works Cited
Itrsquos Acceptablerdquo Pew Research Center (blog) April 16 2020 https wwwpewresearchorgfact-tank20200416most-americans-dont-think-cellphone-tracking-will-help-limit-covid-19-are-divided-on-whether-its- acceptable
Anderson Monica and Andrew Perrin 2017 ldquoDisabled Americans Less Likely to Use Technologyrdquo Pew Research Center (blog) April 7 2017 httpswwwpewresearchorgfact-tank20170407disabled-americans- are-less-likely-to-use-technology
Apple and Google nd ldquoPrivacy-Preserving Contact Tracingrdquo Apple Accessed May 19 2020 httpswwwapplecomcovid19contacttracing
ASTHO 2012 ldquoPublic Health Collection Use Sharing and Protection of Informationrdquo Issue Brief ASTHO Legal Preparedness Series Association of State and Territorial Health Officials httpswwwasthoorgProgramsPreparednessPublic-Health-Emergency-LawPublic-Health-and- Information-Sharing-ToolkitCollection-Use-Sharing-and-Protection- Issue-Brief
Auxier Brooke Lee Rainie Monica Anderson Andrew Perrin Madhu Kumar and Erica Turner 2019 ldquoAmericans and Privacy Concerned Confused and Feeling Lack of Control Over Their Personal Informationrdquo Pew Research Center Internet Science amp Tech (blog) November 15 2019 httpswwwpewresearchorginternet20191115americans-and-privacy- concerned-confused-and-feeling-lack-of-control-over-their-personal-infor mation
Barth Susanne and Menno D T de Jong 2017 ldquoThe Privacy Paradox ndash Inves-tigating Discrepancies between Expressed Privacy Concerns and Actual Online Behavior ndash A Systematic Literature Reviewrdquo Telematics and Infor-matics 34 (7) 1038ndash58 httpsdoiorg101016jtele201704013
Bernstein Justin Taylor A Holroyd Jessica E Atwell Joseph Ali and Rupali J Limaye 2019 ldquoRockland Countyrsquos Proposed Ban against Unvaccinated Minors Balancing Disease Control Trust and Libertyrdquo Vaccine June httpsdoiorg101016jvaccine201905093
Berry Scott M Elizabeth A Petzold Peter Dull Nathan M Thielman Coleen K Cunningham G Ralph Corey Micah T McClain et al 2016 ldquoA Response Adaptive Randomization Platform Trial for Efficient Evalua-tion of Ebola Virus Treatments A Model for Pandemic Responserdquo Clinical Trials (London England) 13 (1) 22ndash30 httpsdoiorg101177 1740774515621721
Beukenhorst AL DM Schultz J McBeth R Lakshminarayana JC Sergeant and WG Dixon 2017 ldquoUsing Smartphones for Research Outside Clinical
Works Cited 129
Settings How Operating Systems App Evelopers and Users Determine Geolocation Data Quality in MHealth Studiesrdquo In MEDINFO 2017 Precision Healthcare through Infomatics 10ndash14 IOS Press httpebooks iospressnlvolumearticle48095
Bradshaw Tim 2020 ldquo2 Billion Phones Cannot Use Google and Apple Con-tact-Tracing Tech | Ars Technicardquo Ars Technica April 20 2020 httpsarstechnicacomtech-policy2020042-billion-phones-cannot-use-google-and-apple-contract-tracing-tech
Canca Cansu 2020 ldquoWhy lsquoMandatory Privacy-Preserving Digital Contact Tracingrsquo Is the Ethical Measure against COVID-19rdquo Medium May 4 2020 httpsmediumcomcansucancawhy-mandatory-privacy- preserving-digital-contact-tracing-is-the-ethical-measure-against- covid-19-a0d143b7c3b6
Cavalier Robert J ed 2011 Approaching Deliberative Democracy Theory and Practice Carnegie Mellon University Press
Cavoukian Ann 2010 ldquoPrivacy by Design The 7 Foundational Principlesrdquo Ontario Canada Information and Privacy Commissioner of Canada httpsiapporgmediapdfresource_centerpbd_implement_7found_ principlespdf
CDC 2005 ldquoVI Plan for Surveillance of Contacts of SARS Case Supplement B SARS Surveillancerdquo Centers for Disease Control and Prevention httpswwwcdcgovsarsguidanceb-surveillancecontactshtml
mdashmdashmdash 2020a ldquoContact Tracingrdquo Get and Keep America Open Supporting States Tribes Localities and Territories Centers for Disease Control and Prevention httpswwwcdcgovcoronavirus2019-ncovphpopen-americacontact-tracinghtml
mdashmdashmdash 2020b ldquoContact Tracing Part of a Multipronged Approach to Fight the COVID-19 Pandemicrdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention httpswwwcdcgovcoronavirus2019- ncovphpprinciples-contact-tracinghtml
mdashmdashmdash 2020c ldquoInterim Guidance for Businesses and Employersrdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention httpswwwcdcgovcoronavirus2019-ncovcommunityguidance-business- responsehtml
mdashmdashmdash 2020d ldquoOperational Considerations for the Identification of Healthcare Workers and Inpatients with Suspected COVID-19 in Non-US Healthcare Settingsrdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention httpswwwcdcgovcoronavirus2019-ncovhcpnon-us-settingsguidance-identify-hcw-patientshtml
130 Works Cited
mdashmdashmdash 2020e ldquoPreliminary Criteria for the Evaluation of Digital Contact Trac-ing Tools for COVID-19rdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention httpswwwcdcgovcoronavirus 2019-ncovdownloadsphpprelim-eval-criteria-digital-contact-tracingpdf
mdashmdashmdash 2020f ldquoLegal Authorities for Isolation and Quarantinerdquo Centers for Disease Control and Prevention httpswwwcdcgovquarantine aboutlawsregulationsquarantineisolationhtml
mdashmdashmdash 2020g ldquoDiscontinuation of Isolation for Persons with COVID-19 Not in Healthcare Settingsrdquo Interim Guidance Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention httpswww cdcgovcoronavirus2019-ncovhcpdisposition-in-home-patientshtml
mdashmdashmdash 2020h ldquoCOVID-19 Provisional CountsndashWeekly Updates by Select Demographic and Geographic Characteristicsrdquo CDC National Vital Statis-tics System httpswwwcdcgovnchsnvssvsrrcovid_weeklyindexhtm
Center for Survey Measurement (CSM) MEMORANDUM FOR Associate Directorate for Research and Methodology (ADRM) 2017 ldquoRespondent Confidentiality Concernsrdquo September 20 2017 httpswww2censusgovcacnacmeetings2017-11Memo-Regarding-Respondent-Confidentiality- Concernspdf
Cheng Hao-Yuan Shu-Wan Jian Ding-Ping Liu Ta-Chou Ng Wan-Ting Huang and Hsien-Ho Lin 2020 ldquoContact Tracing Assessment of COVID-19 Transmission Dynamics in Taiwan and Risk at Different Exposure Periods Before and After Symptom Onsetrdquo JAMA Internal Medicine May httpsdoiorg101001jamainternmed20202020
Cochrane Emily Claire Cain Miller and Jim Tankersley 2020 ldquoTrump Admin-istration Scales Back Paid Leave in Coronavirus Relief Lawrdquo The New York Times April 2 2020 sec US httpswwwnytimescom20200402uspoliticscoronavirus-paid-leavehtml
Crocker Andrew Kurt Opsahl and Bennett Cyphers 2020 ldquoThe Challenge of Proximity Apps For COVID-19 Contact Tracingrdquo Electronic Fron-tier Foundation April 10 2020 httpswwwefforgdeeplinks202004challenge-proximity-apps-covid-19-contact-tracing
Danquah Lisa O Nadia Hasham Matthew MacFarlane Fatu E Conteh Fatoma Momoh Andrew A Tedesco Amara Jambai David A Ross and Helen A Weiss 2019 ldquoUse of a Mobile Application for Ebola Contact Tracing and Monitoring in Northern Sierra Leone A Proof-of-Concept Studyrdquo BMC Infectious Diseases 19 (1) 810 httpsdoiorg101186s12879-019-4354-z
de Jong Bouke C Badou M Gaye Jeroen Luyten Bart van Buitenen Emman-
Works Cited 131
uel Andreacute Conor J Meehan Cian OrsquoSiochain et al 2019 ldquoEthical Considerations for Movement Mapping to Identify Disease Transmission Hotspotsrdquo Emerging Infectious Diseases 25 (7) httpsdoiorg103201eid2507181421
Devakumar Delan Geordan Shannon Sunil S Bhopal and Ibrahim Abu-bakar 2020 ldquoRacism and Discrimination in COVID-19 Responsesrdquo Lancet (London England) 395 (10231) 1194 httpsdoiorg101016S0140-6736(20)30792-3
Dixit Pranav 2020 ldquoIndiarsquos Contact Tracing App Is All But Mandatory So This Programmer Hacked It So That He Always Appears Saferdquo BuzzFeed News May 12 2020 httpswwwbuzzfeednewscomarticlepranavdixitindia-aarogya-setu-hacked
Doerr Megan Christine Suver and John Wilbanks 2016 ldquoDeveloping a Trans-parent Participant-Navigated Electronic Informed Consent for Mobile- Mediated Researchrdquo SSRN Scholarly Paper ID 2769129 Rochester NY Social Science Research Network httpsdoiorg102139ssrn2769129
Dredze Mark Michael J Paul Shane Bergsma and Hieu Tran 2013 ldquoCarmen A Twitter Geolocation System with Applications to Public Healthrdquo AAAI Workshops Workshops at the Twenty-Seventh AAAI Conference on Artifi-cial Intelligence httpswwwaaaiorgocsindexphpWSAAAIW13paperview70856497
Eckhoff Philip A and Andrew J Tatem 2015 ldquoDigital Methods in Epidemi-ology Can Transform Disease Controlrdquo International Health 7 (2) 77ndash78 httpsdoiorg101093inthealthihv013
EEOC 2020 ldquoWhat You Should Know About COVID-19 and the ADA the Rehabilitation Act and Other EEO Lawsrdquo US Equal Employment Opportunity Commission May 7 2020 httpswwweeocgovwyskwhat-you-should-know-about-covid-19-and-ada-rehabilitation-act-and-other-eeo-laws
Electronic Privacy Information Center Testimony to Congress 2020 ldquoEPIC to Congress Establish Privacy Safeguards for Digital Contact Tracingrdquo April 15 2020 httpsepicorgtestimonycongressEPIC-HEC-Contact-Tracing-Apr2020pdf
Elliott Douglas J Ana Kreacic Lorenzo Milans del Bosch and Lisa Quest 2020 ldquoData-Sharing in the Time of Coronavirusrdquo Oliver Wyman Forum Accessed May 19 2020 httpswwwoliverwymanforumcomfuture-of-data2020aprdata-sharing-in-the-time-of-coronavirushtml
Faden Ruth R and Tom L Beauchamp 1986 A History and Theory of Informed Consent Oxford University Press
132 Works Cited
FCC and FTC 2017 ldquoFCC-FTC Consumer Protection Memorandum of Understandingrdquo Federal Communications Commission and Federal Trade Commission httpswwwftcgovsystemfilesdocumentscooperation_agreements151116ftcfcc-moupdf
Ferretti Luca Chris Wymant Michelle Kendall Lele Zhao Anel Nurtay Lucie Abeler-Doumlrner Michael Parker David Bonsall and Christophe Fraser 2020 ldquoQuantifying SARS-CoV-2 Transmission Suggests Epidemic Control with Digital Contact Tracingrdquo Science 368 (6491) httpsdoiorg101126scienceabb6936
Fishkin James S and Peter Laslett eds 2003 Debating Deliberative Democ-racy Blackwell Publisher Ltd httpsonlinelibrarywileycomdoibook 1010029780470690734
Flanagan Mary Daniel C Howe and Helen Nissenbaum 2008 ldquoEmbodying Values in Technology Theory and Practicerdquo In Information Technology and Moral Philosophy edited by Jeroen van den Hoven and John Weckert 322ndash53 Cambridge University Press
Fraccaro Paolo Anna Beukenhorst Matthew Sperrin Simon Harper Jasper Palmier-Claus Shocircn Lewis Sabine N Van der Veer and Niels Peek 2019 ldquoDigital Biomarkers from Geolocation Data in Bipolar Disorder and Schizophrenia A Systematic Reviewrdquo Journal of the American Medical Informatics Association 26 (11) 1412ndash20 httpsdoiorg101093jamiaocz043
Fraser Christophe Lucie Abeler-Doumlrner Luca Ferretti Michael Parker Michelle Kendall and David Bonsall 2020 ldquoDigital Contact Tracing Comparing the Capabilities of Centralised and Decentralised Data Architectures to Effectively Suppress the COVID-19 Epidemic Whilst Maximising Freedom of Movement and Maintaining Privacyrdquo httpsgithubcomBDI-pathogenscovid-19_instant_tracingblobmasterCentralised20and20decen-tralised20systems20for20contact20tracingpdf
FTC 2012 ldquoProtecting Consumer Privacy in an Era of Rapid Changerdquo FTC Report Recommendations for Businesses and Policymakers Federal Trade Commission httpswwwftcgovsitesdefaultfilesdocumentsreports federal-trade-commission-report-protecting-consumer-privacy- era-rapid-change-recommendations120326privacyreportpdf
mdashmdashmdash 2020 ldquoPrivacy amp Data Security Update 2019rdquo Federal Trade Commis-sion httpswwwftcgovsystemfilesdocumentsreportsprivacy-data- security-update-20192019-privacy-data-security-report-508pdf
Furlanello Cesare Stefano Merler Stefano Menegon Sebastiano Mancuso and Gianni Bertiato 2002 ldquoNew WEBGIS Technologies for Geo-Location of
Works Cited 133
Epidemiological Data An Application for the Surveillance of the Risk of Lyme Borreliosis Diseaserdquo Giornale Italiano Di Aritmologia e Cardiosti-molazione Special Issue Proceedings of the 15th Int Congress ldquoThe New Frontiers of Arrhythmiasrdquo 5 (1) 241ndash45
Gan Nectar and David Culver 2020 ldquoChina Is Fighting the Coronavirus with a Digital QR Code Herersquos How It Worksrdquo CNN Business April 16 2020 httpswwwcnncom20200415asiachina-coronavirus-qr-code-intl-hnkindexhtml
Gibson Dustin G Adaeze C Wosu George William Pariyo Saifuddin Ahmed Joseph Ali Alain B Labrique Iqbal Ansary Khan Elizeus Rutebemberwa Meerjady Sabrina Flora and Adnan A Hyder 2019 ldquoEffect of Airtime Incentives on Response and Cooperation Rates in Non-Communicable Disease Interactive Voice Response Surveys Randomised Controlled Trials in Bangladesh and Ugandardquo BMJ Global Health 4 (5) e001604 httpsdoiorg101136bmjgh-2019-001604
Giglio Mike 2020 ldquoWould You Sacrifice Your Privacy to Get Out of Quarantinerdquo The Atlantic April 22 2020 httpswwwtheatlanticcompoliticsarchive202004coronavirus-pandemic-privacy-civil-liberties-911 609172
Guariglia Matthew 2020 ldquoThe Dangers of COVID-19 Surveillance Proposals to the Future of Protestrdquo Electronic Frontier Foundation April 29 2020 httpswwwefforgdeeplinks202004some-covid-19-surveillance-propos als-could-harm-free-speech-after-covid-19
Hadavas Chloe 2020 ldquoHow Effective Are Contact Tracing Appsrdquo Slate Maga-zine May 13 2020 httpsslatecomtechnology202005contact- tracing-apps-less-effective-icelandhtml
Hamilton Isobel Asher 2020 ldquoPoland Made an App That Forces Coronavirus Patients to Take Regular Selfies to Prove Theyrsquore Indoors or Face a Police Visitrdquo Business Insider March 23 2020 httpswwwbusinessinsidercompoland-app-coronavirus-patients-mandaotory-selfie-2020-3
Hargittai Eszter Minh Hao Nguyen Jaelle Fuchs Jonathan Gruber Will Marler Amanda Hunsaker and Goumlkccedile Karaoglu 2020 ldquoCovid-19 Study on Digital Media and the Coronavirus Pandemicrdquo Internet Use and Society Division Institute of Communication and Media Research University of Zurich httpwebuseorgcovid
Hargittai Eszter and Elissa Redmiles 2020 ldquoWill Americans Be Willing to Install COVID-19 Tracking Appsrdquo Scientific American Blog Net-work April 28 2020 httpsblogsscientificamericancomobservationswill-americans-be-willing-to-install-covid-19-tracking-apps
134 Works Cited
Hart Vi Divya Siddarth Bethan Cantrell Lila Tretikov Peter Eckersley John Langford Scott Leibrand et al 2020 ldquoOutpacing the Virus Dig-ital Response to Containing the Spread of COVID-19 While Mitigat-ing Privacy Risksrdquo Whitepaper 5 COVID-19 Rapid Response Impact Initiative Edmond J Safra Center for Ethics httpsdrivegooglecomfiled1vIN2AX-DDNW-S0aHq8xs0RJ2jkR_CckXview
Hemming K T P Haines P J Chilton A J Girling and R J Lilford 2015 ldquoThe Stepped Wedge Cluster Randomised Trial Rationale Design Anal-ysis and Reportingrdquo BMJ 350 (February) httpsdoiorg101136bmjh391
Hendrix Steve and Ruth Eglash 2020 ldquoIsrael Is Using Cellphone Surveil-lance to Warn Citizens You May Already Be Infectedrdquo Washington Post Accessed May 19 2020 httpswwwwashingtonpostcomworldmiddle_eastisrael-is-using-cellphone-surveillance-to-warn-citizens-you-may- already-be-infected2020031968267294-69e7-11ea-b199-3a9799c54 512_storyhtml
Heneghan Carl Jon Brassey and Tom Jefferson 2020 ldquoCOVID-19 What Proportion Are Asymptomaticrdquo Centre for Evidence-Based Medi-cine httpswwwcebmnetcovid-19covid-19-what-proportion-are- asymptomatic
HHS 2019 ldquoPublic Health Emergency Declarationrdquo Public Health Emergency Accessed May 19 2020 httpswwwphegovPreparednesslegalPagesphedeclarationaspx
Hickey Matt 2014 ldquoCarriers Can Now Install Apps On Android Handsets Without Customersrsquo Permissionrdquo Forbes December 1 2014 httpswww forbescomsitesmatthickey20141201carriers-can-now-install- apps-on-android-handsets-without-customers-permission
Hinch Robert Will Probert Anel Nurtay Michelle Kendall Chris Wymant Matthew Hall Katrina Lythgoe et al 2020 ldquoEffective Configurations of a Digital Contact Tracing App A Report to NHSXrdquo fileUsersameliahoodDownloadsReport20-20Effective20App20Configurations20(1)pdf
Ingram David 2020 ldquoApple Google Push Makers of Coronavirus Apps Not to Record User Locationrdquo May 4 2020 NBC News Accessed May 19 2020 httpswwwnbcnewscomtechtech-newscoronavirus-apps-won-t-be-able- record-users-location-apple-n1199586
Iwaya Leonardo H Jane Li Simone Fischer-Huumlbner Rose-Mharie Aringhlfeldt and Leonardo A Martucci 2019 ldquoE-Consent for Data Privacy Consent Management for Mobile Health Technologies in Public Health Surveys and
Works Cited 135
Disease Surveillancerdquo Studies in Health Technology and Informatics 264 (August) 1223ndash27 httpsdoiorg103233SHTI190421
ldquoJoint Statement on Contact Tracingrdquo 2020 April 19 httpscryptobriefingcomwp-contentuploads202004Joint-Statement-from-Researcherspdf
Kahn Gilmor Daniel 2020 ldquoPrinciples for Technology-Assisted Contact- Tracingrdquo White Paper American Civil Liberties Union httpswwwaclu orgreportaclu-white-paper-principles-technology-assisted-contact-tracing
Kim Max S 2020 ldquoSeoulrsquos Radical Experiment in Digital Contact Tracingrdquo The New Yorker April 17 2020 httpswwwnewyorkercomnewsnews-deskseouls-radical-experiment-in-digital-contact-tracing
Kim Nemo 2020 ldquolsquoMore Scary than Coronavirusrsquo South Korearsquos Health Alerts Expose Private Livesrdquo The Guardian March 6 2020 sec World news httpswwwtheguardiancomworld2020mar06more-scary- than-coronavirus-south-koreas-health-alerts-expose-private-lives
Kirzinger Ashley Liz Hamel Cailey Muntildeana Audrey Kearney and Mollyann Brodie 2020 ldquoKFF Health Tracking Poll ndash Late April 2020 Coronavirus Social Distancing and Contact Tracingrdquo Kaiser Family Foundation (blog) April 24 2020 httpswwwkfforgcoronavirus-covid-19issue-briefkff-health-tracking-poll-late-april-2020
Knobel Cory and Geoffrey C Bowker 2011 ldquoComputing Ethics Value in Designrdquo Communications of the ACM 54 (7) 26ndash28 httpsdoiorgdoi10114519657241965735
Lee Ramon Rosa R Cui Kathryn E Muessig Harsha Thirumurthy and Joseph D Tucker 2014 ldquoIncentivizing HIVSTI Testing A Systematic Review of the Literaturerdquo AIDS and Behavior 18 (5) 905ndash12 https doiorg101007s10461-013-0588-8
Leprince-Ringuet Daphne 2020 ldquoContact-Tracing Apps Why the NHS Said No to Apple and Googlersquos Planrdquo ZDNet April 28 2020 httpswwwzdnetcomarticlecontact-tracing-apps-why-the-nhs-said-no-to-apple-and- googles-plan
Lovejoy Ben 2020 ldquoNHS Contact Tracing App Hands-on as the Test Goes Liverdquo 9to5Mac (blog) May 7 2020 https9to5maccom20200507nhs-contact-tracing
Mahmood Sultan Khaled Hasan Michelle Colder Carras and Alain Labri-que 2020 ldquoGlobal Preparedness Against COVID-19 We Must Leverage the Power of Digital Healthrdquo JMIR Public Health and Surveillance 6 (2) e18980 httpsdoiorg10219618980
Mathews Simon C Michael J McShea Casey L Hanley Alan Ravitz Alain B Labrique and Adam B Cohen 2019 ldquoDigital Health A Path to
136 Works Cited
Validationrdquo Npj Digital Medicine 2 (1) 1ndash9 httpsdoiorg101038s41746-019-0111-3
Mello By Michelle M and C Jason Wang 2020 ldquoEthics and Governance for Digital Disease Surveillancerdquo Science May httpsdoiorg101126science abb9045
Mills Rodrigo Chris 2020 ldquoDigital Contact Tracing Is Becoming Available but Is It Effectiverdquo The Hill May 7 2020 httpsthehillcompolicytechnolo-gy496498-digital-contact-tracing-is-becoming-available-but-is-it-effective
MIT nd ldquoPrivate Kit Safe Paths Privacy-by-Design Covid19 Solutions Using GPS+Bluetooth for Citizens and Public Health Officialsrdquo Safepaths Accessed May 20 2020 httpssafepathsmitedu
ldquoMobile Cellular Subscriptions (per 100 People)rdquo 2018 World Telecommuni-cationICT Development Report International Telecommunication Union httpsdataworldbankorgindicatorITCELSETSP2
Moore Sarah Anne-Marie Tasseacute Adrian Thorogood Ingrid Winship Marsquon Zawati and Megan Doerr 2017 ldquoConsent Processes for Mobile App Mediated Research Systematic Reviewrdquo JMIR MHealth and UHealth 5 (8) e126 httpsdoiorg102196mhealth7014
Morse Jack 2020 ldquoNorth Dakota Launched a Contact-Tracing App Itrsquos Not Going Wellrdquo Mashable Accessed May 19 2020 httpsmashablecomarticlenorth-dakota-contact-tracing-app
Muller Robert T 2020 ldquoCOVID-19 Brings a Pandemic of Conspiracy Theo-riesrdquo Psychology Today April 24 2020 httpswwwpsychologytodaycomblogtalking-about-trauma202004covid-19-brings-pandemic-conspiracy- theories
Mulligan Stephen P and Chris D Linebaugh 2019 ldquoData Protection Law An Overviewrdquo R45631 Congressional Research Service httpscrsreports congressgovproductpdfRR45631
National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research 1979 ldquoThe Belmont Report Ethical Principles and Guidelines for the Protection of Human Subjects of Researchrdquo US Dept of Health amp Human Services Office of Human Research Protections httpswwwhhsgovohrpregulations-and-policybelmont-reportread-the-belmont-reportindexhtml
NDDoH 2020 ldquoNorth Dakota Launches Care19 App to Combat COVID-19rdquo North Dakota Department of Health April 7 2020 httpswwwhealth ndgovnewsnorth-dakota-launches-care19-app-combat-covid-19
NYC DOH 2020 ldquoAge-Adjusted Rates of Lab Confirmed COVID-19 Non-hospitalized Cases Estimated Non-Fatal Hospitalized Cases and Persons Known to Have Died per 100000 by RaceEthnicity Grouprdquo New York
Works Cited 137
City Department of Health httpswww1nycgovassetsdohdownloadspdfimmcovid-19-deaths-race-ethnicity-04242020-1pdf
OrsquoRielly Michael 2018 ldquoFCC Regulatory Free Arenardquo Federal Communica-tions Commission June 1 2018 httpswwwfccgovnews-eventsblog 20180601fcc-regulatory-free-arena
OSHA 2020 ldquoGuidance on Preparing Workplaces for COVID-19rdquo OSHA 3990-03 2020 Department of Labor Occupational Safety and Health Act of 1970 httpswwwoshagovPublicationsOSHA3990pdf
Pallmann Philip Alun W Bedding Babak Choodari-Oskooei Munyaradzi Dimairo Laura Flight Lisa V Hampson Jane Holmes et al 2018 ldquoAdaptive Designs in Clinical Trials Why Use Them and How to Run and Report Themrdquo BMC Medicine 16 (1) 29 httpsdoiorg101186s12916-018-1017-7
Pew Research Center 2017 ldquoUS Muslims Concerned About Their Place in Society but Continue to Believe in the American Dreamrdquo Religion amp Public Life Pew Research Center httpswwwpewforumorg20170726findings-from-pew-research-centers-2017-survey-of-us-muslims
mdashmdashmdash 2020 ldquoDemographics of Mobile Device Ownership and Adoption in the United Statesrdquo Internet amp Technology Pew Research Center Accessed May 19 2020 httpswwwpewresearchorginternetfact-sheetmobile
PHLS 2002 ldquoPrinciples of the Ethical Practice of Public Healthrdquo Public Health Leadership Society httpswwwaphaorg-mediafilespdfmembergroupsethicsethics_brochureashx
PIH 2020a ldquoPart I Testing Contact Tracing and Community Management of COVID-19rdquo PIH Guide | COVID-19 STOP COVID Partners in Health httpswwwpihorgsitesdefaultfiles2020-04PIH_Guide_COVID_Part_I_Testing_Tracing_Community_Managment_4_4pdf
mdashmdashmdash 2020b ldquoBox It Inrdquo PIH Guide | COVID-19 Partners in Health httpspreventepidemicsorgcovid19resourcesbox-it-in
Resolve to Save Lives 2020 ldquoCOVID-19 Contact Tracing Playbookrdquo Vital Strategies httpscontacttracingplaybookresolvetosavelivesorg
Reston Maeve Kristina Sgueglia and Cheri Mossburg 2020 ldquoGovernors on East and West Coasts Form Pacts to Decide When to Reopen Economiesrdquo CNN Politics April 13 2020 httpswwwcnncom20200413politicsstates-band-together-reopening-plansindexhtml
Rithalia Amber Catriona McDaid Sara Suekarran Lindsey Myers and Amanda Sowden 2009 ldquoImpact of Presumed Consent for Organ Donation on Donation Rates A Systematic Reviewrdquo BMJ 338 (January) httpsdoiorg101136bmja3162
Rodrigues Rafaela Alina Husain Amanda Couture-Carron Leslye E Orloff
138 Works Cited
and Nawal H Ammar 2018 ldquoPromoting Access to Justice for Immigrant and Limited English Proficient Crime Victims in an Age of Increased Immi-gration Enforcement Initial Report from a 2017 National Surveyrdquo Wash-ington DC National Immigrant Womenrsquos Advocacy Project American University Washington College of Law httpniwaplibrarywclamericaneduwp-contentuploadsImmigrant-Access-to-Justice-National-Reportpdf
Simpson Erin and Adam Conner 2020 ldquoDigital Contact Tracing to Contain the Coronavirusrdquo Center for American Progress httpswww americanprogressorgissuestechnology-policynews20200422483521digital-contact-tracing-contain-coronavirus
Singer Eleanor and Cong Ye 2013 ldquoThe Use and Effects of Incentives in Surveysrdquo The ANNALS of the American Academy of Political and Social Science 645 (1) 112ndash41 httpsdoiorg1011770002716212458082
Thornton Rebecca L 2008 ldquoThe Demand for and Impact of Learning HIV Statusrdquo The American Economic Review 98 (5) 1829ndash63 httpsdoi org101257aer9851829
US DOE 2009 ldquoState Regulation of Private Schoolsrdquo US Dept of Education Office of Non-Public Education httpswww2edgovadminscommchoiceregprivschlregprivschlpdf
Valentino-DeVries Jennifer 2020 ldquoCellphone Carriers Face $200 Million Fine for Not Protecting Location Datardquo The New York Times February 28 2020 sec Technology httpswwwnytimescom20200228technologyfcc-cellphones-location-data-fineshtml
Valentino-DeVries Jennifer Natasha Singer and Aaron Krolik 2020 ldquoA Scramble for Virus Apps That Do No Harmrdquo The New York Times April 29 2020 sec Business httpswwwnytimescom20200429businesscoronavirus-cellphone-apps-contact-tracinghtml
Washington PostndashUMD ldquoWashington Post-University of Maryland National Pollrdquo 2020 Washington Post April 21 2020 httpswwwwashingtonpostcomcontextwashington-post-university-of-maryland-national-poll-april-21-26-20203583b4e9-66be-4ed6-a457-f6630a550ddf
Watson Crystal Anita Cicero James Blumenstock and Michael Fraser 2020 ldquoA National Plan to Enable Comprehensive COVID-19 Case Finding and Contact Tracing in the USrdquo Johns Hopkins Bloomberg School of Public Health Center for Health Security and the Association of State and Terri-torial Health Officials httpswwwcenterforhealthsecurityorgour-workpubs_archivepubs-pdfs2020200410-national-plan-to-contact-tracingpdf
Works Cited 139
WHO 2017 ldquoWHO Guidelines on Ethical Issues in Public Health Surveillancerdquo World Health Organization httpswwwwhointethicspublicationspublic-health-surveillanceen
mdashmdashmdash 2020 ldquoCoronavirus Disease 2019 (COVID-19)rdquo Situation Report 73 World Health Organization httpswwwwhointdocsdefault-sourcecoro-navirusesituation-reports20200402-sitrep-73-covid-19pdf
- _iy90yrtqgtxz
- _GoBack
- Acknowledgments
- Preface
- Lead Authors and Contributors
- Acronyms and Abbreviations
- Summary
-
- Introduction
- DCTT Features Functions and Potential Applications
- Summary of Recommendations
- Summary of Analysis
-
- Introduction
-
- Guiding Principles for the Use of Digital Public Health Technologies for Pandemic Response
-
- Public Health Perspective
-
- Types of Information Collected through Contact Tracing
- How Contact Tracing Information Informs Public Health Action
- Characteristics That Make Data Useful to Public Health for Reducing Disease Transmission
-
- Digital Technology and Contact Tracing
-
- Characteristics of SARS-CoV-2 Relevant to Candidate Digital Solutions
- Previously Existing Contact Tracing Technologies
- Introduction of Novel Digital Contact Tracing Technologies
- Relevant Differences between Manual and Digital Contact Tracing
-
- Ethics of Designing and Using DCTT
-
- Justifying the Use of DCTT Systems
- Monitoring and Evaluating Technologies to Inform Policy and Practice
- Public Trust and Public Attitudes
- Designing Flexible Technology to Maximize Public Health Utility While Respecting Other Values
- Policy Positions to Advance Widespread Use of Digital Contact Tracing Technologies
- Disclosure and AuthorizationConsent
- Promoting Equity and Fairness
- Instituting Transparent Governance and Oversight
-
- Legal Considerations
-
- Data Privacy and Data Security Laws
- Health Information Privacy
- Labor and Employment Privacy Rights
- Constitutional Privacy Rights
- Consent
- Anti-discrimination and Individual Freedom Laws
-
- Recommendations
-
- Public Health
- Ethics
- Legislative
-
- Resources
-
- US Government Response
- Other Governmental and Nongovernmental Organizations
- Digital Contact Tracing Experiences from Other Countries
- Specific Digital ProductsApps
- Polling
- Popular Press
- Commentaries
- Academic Literature
-
- Works Cited
-
copy 2020 Johns Hopkins University Press
This work is also available in an Open Access edition which is licensed
under a Creative Commons AttributionndashNonCommercialndashNoDerivatives 40
International License httpscreativecommonsorglicensesby-nc-nd40
All rights reserved Published 2020
Printed in the United States of America on acid-free paper
9 8 7 6 5 4 3 2 1
Johns Hopkins University Press
2715 North Charles Street
Baltimore Maryland 21218-4363
wwwpressjhuedu
Library of Congress Cataloging-in-Publication Data is available
ISBN 978-1-4214-4061-3 (paperback acid-free paper)
ISBN 978-1-4214-4062-0 (ebook)
ISBN 978-1-4214-4063-7 (ebook open access)
Special discounts are available for bulk purchases of this book For more information
please contact Special Sales at specialsalespressjhuedu
Johns Hopkins University Press uses environmentally friendly book materials
including recycled text paper that is composed of at least 30 percent post-
consumer waste whenever possible
Contents
Lead Authors and Contributors vii
Preface ix
Acknowledgments xi
Acronyms and Abbreviations xiii
Summary 1
Introduction 1
DCTT Features Functions and Potential Applications 3
Summary of Recommendations 9
Summary of Analysis 13
Introduction 23
Guiding Principles for the Use of Digital Public Health
Technologies for Pandemic Response 25
1 Public Health Perspective 29
Types of Information Collected through Contact Tracing 29
How Contact Tracing Information Informs Public Health Action 31
Characteristics That Make Data Useful to
Public Health for Reducing Disease Transmission 32
2 Digital Technology and Contact Tracing 35
Characteristics of SARS-CoV-2 Relevant to Candidate Digital Solutions 35
Previously Existing Contact Tracing Technologies 36
Introduction of Novel Digital Contact Tracing Technologies 37
Relevant Differences between Manual and Digital Contact Tracing 41
vi Contents
3 Ethics of Designing and Using DCTT 43
Justifying the Use of DCTT Systems 44
Monitoring and Evaluating Technologies to Inform Policy and Practice 45
Public Trust and Public Attitudes 48
Designing Flexible Technology to Maximize Public Health Utility
While Respecting Other Values 50
Policy Positions to Advance Widespread Use of Digital Contact
Tracing Technologies 59
Disclosure and AuthorizationConsent 63
Promoting Equity and Fairness 69
Instituting Transparent Governance and Oversight 72
4 Legal Considerations 75
Data Privacy and Data Security Laws 77
Health Information Privacy 82
Labor and Employment Privacy Rights 84
Constitutional Privacy Rights 86
Consent 93
Anti-discrimination and Individual Freedom Laws 94
5 Recommendations 97
Public Health 97
Ethics 98
Legislative 102
Resources 103
US Government Response 103
Other Governmental and Nongovernmental Organizations 106
Digital Contact Tracing Experiences from Other Countries 112
Specific Digital ProductsApps 115
Polling 117
Popular Press 119
Commentaries 120
Academic Literature 121
Works Cited 127
Lead Authors and Contributors
vii
Lead Authors
Joseph Ali JD Assistant Professor Department of International Health Johns Hopkins Bloomberg School of Public Health Core Faculty amp Associate Director for Global Programs Johns Hopkins Berman Institute of Bioethics
Anne Barnhill PhD Core Faculty amp Research Scholar Johns Hopkins Berman Institute of Bioethics
Anita Cicero JD Deputy Director Johns Hopkins Center for Health Security Visiting Faculty Johns Hopkins Bloomberg School of Public Health
Katelyn Esmonde PhD Hecht-Levi Postdoctoral Fellow Johns Hopkins Berman Institute of Bioethics
Amelia Hood MA Research Program Coordinator Johns Hopkins Berman Insti-tute of Bioethics
Brian Hutler PhD JD Hecht-Levi Postdoctoral Fellow Johns Hopkins Berman Institute of Bioethics
Jeffrey Kahn PhD MPH Andreas C Dracopoulos Director Johns Hopkins Ber-man Institute of Bioethics
Alan Regenberg MBE Director of Outreach amp Research Support Associate Fac-ulty Johns Hopkins Berman Institute of Bioethics
Crystal Watson DrPH MPH Senior Scholar Johns Hopkins Center for Health Security Assistant Professor Department of Environmental Health and Engi-neering Johns Hopkins Bloomberg School of Public Health
Matthew Watson Senior Analyst Johns Hopkins Center for Health Security Senior Research Associate Department of Environmental Health and Engi-neering Johns Hopkins Bloomberg School of Public Health
Other Contributors
Robert Califf MD MACC Head of Clinical Policy and Strategy Verily and Goo-gle Health
Ruth Faden PhD MPH Philip Franklin Wagley Professor of Biomedical Ethics amp Founder Johns Hopkins Berman Institute of Bioethics
viii Lead Authors and Contributors
Divya Hosangadi MSPH Senior Analyst Johns Hopkins Center for Health Secu-rity Research Associate Department of Environmental Health and Engi-neering Johns Hopkins Bloomberg School of Public Health
Nancy Kass ScD Deputy Director for Public Health amp Phoebe R Berman Pro-fessor of Bioethics and Public Health Johns Hopkins Berman Institute of Bioethics
Alain Labrique PhD MHS MS Professor Department of International Health Johns Hopkins Bloomberg School of Public Health Director Johns Hopkins Global Health Initiative
Deven McGraw JD MPH LLM Chief Regulatory Officer CiitizenMichelle Mello JD PhD Professor of Law Stanford Law School Professor of
Health Research and Policy Stanford University School of MedicineMichael Parker BEd (Hons) MA PhD Director Wellcome Centre for Ethics and
Humanities Ethox Centre University of OxfordStephen Ruckman JD MSc MAR Senior Advisor to the President for Policy
Office of the President Johns Hopkins UniversityLainie Rutkow JD MPH PhD Senior Advisor to the President for National Cap-
ital Academic Strategy Office of the President Johns Hopkins UniversityJosh Sharfstein MD Vice Dean for Public Health Practice and Community
Engagement Professor of the Practice Johns Hopkins Bloomberg School of Public Health
Jeremy Sugarman MD MPH MA Deputy Director for Medicine Harvey M Meyerhoff Professor of Bioethics and Medicine Johns Hopkins Berman Institute of Bioethics Department of Medicine Johns Hopkins School of Medicine and Department of Health Policy and Management Johns Hop-kins Bloomberg School of Public Health
Eric Toner MD Senior Scholar Johns Hopkins Center for Health Security Senior Scientist Department of Environmental Health and Engineering Johns Hop-kins Bloomberg School of Public Health
Marc Trotochaud MSPH Analyst Johns Hopkins Center for Health Security Research Associate Department of Environmental Health and Engineering Johns Hopkins Bloomberg School of Public Health
Effy Vayena PhD Professor Health Ethics amp Policy Lab Department of Health Sciences amp Technology ETH Zurich
Tal Zarsky JSD LLM LLB Professor of Law University of Haifa Faculty of Law Visiting Scholar University of Pennsylvania Law School (2019ndash2020)
Preface
ix
Digital technologies are being developed and promoted to support the public health response to the COVID-19 pandemic with discussion and implementation planning in the United States by localities states institu-tions and employers Key decision makers and stakeholdersmdashincluding government officials institutional leaders employers digital technology developers and the publicmdashrequire clear and well-supported guidance to inform the deployment and use of these technologies as well as of the data they collect store and share While technology-based approaches are currently unable to provide solutions on their own experiences in other countries indicate that they could be used successfully in conjunc-tion with traditional and novel public health methods
This report reflects a rapid research and expert consensus group ef-fort led by the Berman Institute of Bioethics and the Center for Health Se-curity at Johns Hopkins University It draws on experts from both inside and outside Johns Hopkins in bioethics health security public health technology development engineering public policy and law The report highlights issues that must be addressed and provides recommendations for the use of digital technologies as part of contact tracing
The analysis offered here is focused on answering the following questions
bull Can digital contact tracing technologies (DCTT) be effective as part of public health responses to the pandemic and if so to what degree for which specific types of functions with what confidence and with what requirements
bull How can these technologies serve the interests of public health while respecting other individual and collective interests such as ensuring equitable distribution of benefits and burdens and limit-ing infringement on privacy and other civil liberties
x Preface
bull What are the ethical legal policy and governance guardrails cur-rently in place around such technologies and what else is needed
bull What additional guardrails are required to ensure that the goals of public health in using these technologies are achievable in ways that are ethically and legally sound
To answer these questions the report examines some core aspects of dig-ital technologies applied to contact tracing focusing on
bull the value of and basic methods for traditional public health surveil-lance and contact tracing
bull candidate technological products to enhance public health surveil-lance and contact tracing how they work and their comparative value for public health
bull core ethical legal and governance considerations and how they relate to relevant features of candidate technological solutions and
bull what is needed to move forward responsibly with the use of digital technology in support of public health surveillance acknowledg-ing gaps in our current understanding
The project involved in-depth analysis by a dedicated team of faculty postdoctoral fellows and research staff working over the course of only a few weeks but with great intensity drafting a report in collaboration with 26 total contributors writing commenting and revising through multiple drafts with the penultimate draft ldquopressure-testedrdquo by review and discus-sion at a virtual workshop of invited experts and stakeholders held on May 13 2020 and the final version completed on May 21 2020 The report and analysis builds on the excellent work of others in some parts of this territory while focusing on the gaps in analysis and areas that have not been sufficiently addressed The goal is to offer comprehensive guid-ance to relevant stakeholders to advance public health response during the COVID-19 pandemic Given the rapidly evolving territory into which DCTT is being introduced this report will by necessity be something of a living document updated as often as information dictates in order to continue to offer leading-edge analysis and guidance Versions will be noted in the digital and print editions
Acknowledgments
xi
Efforts like this project require teams and even small armies to be carried out successfully and this was no exception except that it was many fewer people working many more hours than could reasonably be expected of them From the initial kernel of an idea to the publication of this report in book form this project took just over a month total That seems impossi-ble even as I know it is accurate and it speaks to the incredible commit-ment hard work research skills and analytic acumen of our colleagues at Johns Hopkinsmdashthe core team are deservedly listed as lead authors of this report
None of this would have been possible without the supportmdashmoral and financialmdashand encouragement of Johns Hopkins University Presi-dent Ronald J Daniels who was the first to suggest the idea to me of taking on this topic He provided not only support and encouragement but the imprimatur of his office including help guidance and counsel from Prof Lainie Rutkow senior advisor to the president Lainie played a more integral role than that description captures reflected in part by her inclusion among the reportrsquos contributors but she deserves special acknowledgement for shepherding us through to the end
I mentioned that this was a team effort and every team requires an effective leader My colleague Prof Joseph Ali stepped into that role as we undertook the project and then he worked seven days a week along with the rest of the core research and writing team always unfailingly positive and deeply engaged in the work He along with Prof Anne Barnhill Alan Regenberg Amelia Hood and Drs Katelyn Esmonde Brian Hutler and Crystal Watson all deserve special thanks for doing so much in so little time all while working under the grinding social distancing restrictions of the 2020 pandemic That work was supported by Arnold amp Porter Kaye Scholer LLP with legal research and other assistancemdasha huge thanks to
xii Acknowledgments
them Finally the 16 contributing authors were incredibly generous with their time energy and insights all on ridiculously tight timelines and never a complaint or objection
The project benefited greatly from a number of experts who provided written feedback on drafts and who attended the virtual workshop to test our recommendations including Miles Stewart Rob Nichols Smisha Aagarwal Karl Steiner Anupam Joshi Charles Scheeler Ford Rowan and Jay Wagley
Last the fact that this report appears in published book form by Johns Hopkins University Press is another minor miracle from manu-script to printed book in under a week Thanks to JHUP Director Barbara Kline Pope and her team for being willing to take on the challenge and for the incredible focused effort it required
my heartfelt thanks andappreciation to you all
Jeffrey Kahn
Acronyms and Abbreviations
xiii
ADA Americans with Disabilities ActBLE Bluetooth Low EnergyCalOPPA California Online Privacy Protection ActCBP Customs and Border ProtectionCCPA California Consumer Privacy ActCDC Centers for Disease Control and PreventionCLOUD Act Clarifying Lawful Overseas Use of Data ActCOPPA Childrenrsquos Online Privacy Protection ActCOV+ confirmed positive SARS-CoV-2 test resultCOVID-19 coronavirus disease 2019CPNI customer proprietary network informationCSLI cell-site location informationDCTT digital contact tracing technology and closely related
digital health productsECPA Electronic Communications Privacy ActEEOC Equal Employment Opportunity CommissionE-SIGN Electronic Signatures in Global and National
Commerce ActEU European UnionFCC Federal Communications CommissionFTC Federal Trade CommissionFTCA Federal Trade Commission ActGIS geographic information systemGPS global positioning systemHHS US Department of Health and Human ServicesHIPAA Health Insurance Portability and Accountability ActHIV human immunodeficiency virusICU intensive care unit
xiv Acronyms and Abbreviations
IRB institutional review boardJHU Johns Hopkins UniversityOCR Office for Civil Rights US Department of Health amp
Human ServicesOSHA Occupational Safety and Health AdministrationPHI protected health informationPII personally identifiable informationPPE personal protective equipmentPPPT privacy-preserving proximity trackingQR Code quick response codeRFRA Religious Freedom Restoration ActRLUIPA Religious Land Use and Institutionalized Persons ActSARS-CoV-2 severe acute respiratory syndrome coronavirus 2SCA Stored Communications ActSTI sexually transmitted infection
DIGITAL CONTACT TRACING FOR PANDEMIC RESPONSE
11
Summary
Introduction
Public health professionals around the world are working tirelessly to respond to the COVID-19 pandemic using tried-and-tested public health methods for infectious disease surveillance and control These traditional methods are essential to the global COVID-19 response To complement these actions and potentially augment the speed and efficacy of the pub-lic health workforce digital technologies are being harnessed Given the scale of the pandemic significant efforts are being undertaken to develop and leverage public-facing and health-system-supportive technology solu-tions including smartphone apps and other digital tools that may aid public health surveillance and contact tracing
Digital contact tracing technology and closely related digital health products (together DCTT) have been used in several countries as part of broader disease surveillance and containment strategies In the United States DCTT has been proposed as an integral part of some plans to ldquoreopenrdquo the country (Allen et al 2020 Hart et al 2020 Simpson and Conner 2020) It is almost certain that these and related technologies will become part of not only the COVID-19 response but also the larger toolbox for future public health communicable disease prevention and control
These technologies have significant promise They also raise import-ant ethical legal and governance challenges that require comprehensive analysis in order to support decision-making Government officials pub-lic health leaders leaders of institutions employers digital technology de-velopers and the public all must be adequately informed in order to make
2 Digital Contact Tracing for Pandemic Response
responsible choices Johns Hopkins University recognized the importance of helping to guide this process It organized an expert group with mem-bers from inside and outside of Hopkins and led by its Berman Institute of Bioethics in collaboration with the Center for Health Security Its charge was to examine the ethics law policy and public health implications of using digital technologies as part of pandemic response and to develop guidance including a framework and actionable recommendations for governmental and institutional decision makers
Overall this expert group urges a stepwise approach that prioritizes align-ment of technology with public health needs and public values building choice into design architecture and capturing real-world results and impacts to allow adjustments as required Further we urge an approach that recog-nizes that there are complicated issues to resolve for governments insti-tutions and businesses and that introduction of DCTT must include public engagement and ongoing assessments to improve both performance and adoption
Specific recommendations include the following
bull There is no ldquoone size fits allrdquo approach to DCTT Technology design should not be static but it should be capable of evolving depending upon local conditions new evidence and changing preferences and priorities
bull Technology companies alone should not control the terms condi-tions or capabilities of DCTT nor should they presume to know what is acceptable to members of the public
bull DCTT should be designed to have a base set of features that pro-tect privacy with layers of additional capabilities that users may choose to activate An initial default should be that user location data are not shared but users should be provided with easy mech-anisms and prompts to allow for opting-in to this capability with encouragement to the public if it is shown to be critical to achieving public health goals
bull Data collected through DCTT should be made available to public health professionals and to researchers in de-identified form to support population-level epidemiologic analyses
Summary 3
bull Those who authorize use of DCTT within a particular jurisdic-tion or institution should continuously and systematically monitor the technologyrsquos performance in that context This should include monitoring for effectiveness and benefit monitoring for harms and monitoring for the fair distribution of both benefits and harms
bull Governments should not require mandatory use of DCTT given uncertainty about potential burdens and benefits Additional tech-nology user and real-world testing is needed
Through in-depth analysis and recommendations this report seeks to guide decision-making and enhance understanding of
bull the value of and basic methods for traditional public health surveil-lance and contact tracing
bull candidate technological products to enhance public health surveil-lance and contact tracing and their comparative value for public health
bull core ethical legal policy and governance considerations and how they relate to relevant features of candidate technological solutions and
bull what is needed to move forward responsibly with the use of digital technology in support of public health surveillance acknowledg-ing gaps in our current understanding
The full set of recommendations are intended to (1) support effective and informed adoption of DCTT (2) encourage design of flexible technol-ogies that maximize public health utility while respecting other values (3) establish meaningful processes for user disclosure and authorization (consent) (4) promote equity and fairness in the uses of DCTT and (5) foster transparent governance and oversight
DCTT Features Functions and Potential Applications
Digital contact tracing technologies and platforms can be roughly catego-rized into three broad approaches along a spectrum of potential policies and methods a maximal approach (typified by the South Korean govern-
4 Digital Contact Tracing for Pandemic Response
mentrsquos centralized and triangulated data collection (M S Kim 2020)) a minimal approach (typified by the AppleGoogle decentralized priva-cy-preserving proximity tracking (PPPT) and contact notification (Apple and Google nd)) and a diverse range of middle-ground approaches that aim to augment manual contact tracing with the collection of digital data that can be shared with public health authorities
Minimal approaches such as the AppleGoogle PPPT use Bluetooth Low Energy (BLE) ldquohandshakesrdquo that record close contact between mo-bile phone users but do not register the location in which the contact hap-pened In most architectures these proximity data are stored in the usersrsquo phones as anonymized ldquobeaconsrdquo that cannot be used to re-identify the users directly If a user with a PPPT app installed on their phone tests pos-itive and enters test results into their app those who have been in contact with them can be notified by the app This ldquoexposure notificationrdquo can be automatic or at the discretion of the COV+ person depending on the app design If notified a user who has been in contact with a COV+ individ-ual would receive a push notification alerting them to possible exposure (which may be timestamped) but with no other identifying information
The most prevalent middle-ground approach in the US context in-volves the collection and storage of personal datamdashincluding identifying information and location datamdashon the userrsquos phone These decentralized but personally identifiable data can then be voluntarily shared with pub-lic health officials if the user tests positive for SARS-CoV-2 (severe acute respiratory syndrome coronavirus 2) For example a team at the Mas-sachusetts Institute of Technology (MIT) has developed an app called Private Kit Safe Paths (MIT nd) that stores usersrsquo location data on their phone for 28 days If a user tests positive she can voluntarily upload her location data to a website that is accessible only to public health officials Officials can then analyze these personally identifiable data and subse-quently broadcast redacted and de-identified data to other users Healthy users would have access to these redacted location data of COV+ users but their own data would not leave their phones At a minimum the storage of user location data can function as a ldquomemory aidrdquo if the user tests positive but releasing the data to public health authorities may help to analyze the spread of SARS-CoV-2 and alert individuals or groups that have been in contact with COV+ patients
Summary 5
The US Centers for Disease Control and Prevention (CDC) has pub-lished preliminary criteria for evaluating capabilities and attributes of DCTT (CDC 2020e) These and other resources suggest that a compre-hensive assessment of DCTT and its potential to advance the publicrsquos health will require careful consideration of numerous interconnected fac-tors that interact in complex ways and must be navigated within the chal-lenging contexts of uncertainty and urgent need (Figure 1) These include
bull scientific and epidemiological understanding of SARS-CoV-2 transmission and infection
bull public health needs for combating the outbreak
bull technological capabilities of DCTT
bull performance of DCTT applications
bull ethical values and principles
bull characteristics of public adoption and acceptance and
bull legal issues and landscape
FIGURE 1 Interrelating Factors That Frame Responsible Development of Digital Contact
Tracing Technology
6 Digital Contact Tracing for Pandemic Response
The primary objectives for use of DCTT during the COVID-19 pandemic must be to reduce illness and death and facilitate public health efforts to reduce transmission of the virus These objectives fall under a broader overall goal of contributing to societal well-being during the pandemic It is not yet known whether and how much DCTT can contribute to these primary objectives nor whether it will be able to contribute without gen-erating new burdens or even harms such as incorrect warnings or ldquonoiserdquo that detract from the work of manual contact tracing
The process of identifying acceptable technology designs and uses is complex given the interplay among the factors Our analysis reveals that there is no ldquoone size fits allrdquo approach to DCTT There is variability across the United States with respect to SARS-CoV-2 prevalence and in-fection rates public health capacity public attitudes toward DCTT and acceptability of various potential features Moreover our understanding of SARS-CoV-2 and DCTT is evolving public health response needs and capabilities are changing and public attitudes are shifting Different tech-nologies used in different ways may be appropriate to achieve slightly dif-ferent public health goals in different localities and at different points in the pandemic A tiered and phased approach to technology development should be facilitated by law and policy prioritizing underlying interoper-ability while permitting user choices now and for the future
Given the complexity of the terrain as a first step those developing or considering widespread use of DCTT as part of pandemic response should be guided by the following principles and related actions (see box) These principles are meant to apply to DCTT as well as other dig-ital technologies used in novel ways during pandemic response
These principles make clear that in order to maximize the public good from use of DCTT public health needs and technological capabili-ties must be carefully aligned Government officials public health leaders leaders of other institutions employers digital technology developers and the public are all key stakeholders that must be informed and en-gaged in order to enable the most successful and ethically acceptable uses of DCTT
Guiding Principles for the Use of Digital Public Health Technologies for Pandemic Response
Transparency and public engagement are essential to an inclusive digital public health response
bull Government public health and digital technology leaders must engage
effectively with the public and other stakeholders to communicate the util-
ity importance oversight and limitations of relevant digital technologies
including their implications for individualsrsquo privacy and civil liberties
bull Transparency at all levels is essential for maintaining public trust and
confidence
bull To the extent possible digital public health responses should reflect the
range of values that are important to individuals including advancing the
health and well-being of the community as a whole
bull Decision makers should recognize the sacrifices that some people may be
willing to make during a pandemic in order to advance public health goals
Acceptance by some of particular monitoring capabilities should not be read
as a willingness to extend these methods to other problems or uses
Digital public health responses must represent the least infringement of civil liberties necessary to accomplish the public health goals
bull If preferred digital public health strategies infringe on privacy and other civil
liberties the infringements must be sufficiently justified by the circum-
stances of the pandemic offset by ample anticipated public benefit and
considered relative to infringements associated with other possible strate-
gies such as mass physical distancing
bull Only those data that are necessary and relevant for the stated public health
purposes should be collected Identifiable data should be stored in a secure
manner and only for the period of time that the public health purposes
require
bull Adopted technologies should not be used in ways that subject communities
to discrimination or surveillance for nonndashpublic health reasons
bull Respect for individual autonomy requires that users are sufficiently informed
of the public health goals of the technology and the extent to which those
goals are being met
Use of digital public health technologies and data must be guided by best available evidence
bull Decisions to deploy digital public health technologies should be based on a
careful assessment of the uses and limitations of any proposed technology
taking into account the best available evidence
bull Those who deploy digital public health technologies should continuously and
systematically monitor their performance as well as any evidence that is being
generated in other contexts about the selected technological solution and about
other competing technologies
bull Unintended consequencesmdashincluding those that might impact public health
goals core values and interests of the public and unfair advantage or disadvan-
tagemdashshould be carefully monitored and addressed as necessary
Responsible use of digital public health technology requires meaningful governance and accountability
bull Systems of governance must be trustworthy and well informed They must be
reviewed and adjusted as circumstances and evidence change or as unintended
effects are identified
bull Trusted representatives who are capable of developing and implementing uni-
form and fair standards for adopting and utilizing underlying digital technology
must be identified
bull Understandable transparent and publicly accessible rules must guide the
collection access control use storage and combination of data by government
authorities public and private institutions and other parties such as public
health researchers
bull Oversight accountability and consequences for abuse or misuse of these data
must be explicit and enforceable
The deployment of digital public health technology must be rooted in a commitment to equity
bull Digital public health technologies should be deployed in a manner that does
not propagate preexisting patterns of unfair disadvantage or further distribute
harms and risks unfairly throughout the population
bull To the extent possible digital public health technologies should be designed to
rectify existing inequities
bull Oversight mechanisms must be in place to ensure that the improved public
health outcomes are equitable and to detect and correct any unforeseen resul-
tant injustices attributable to the technology or that can be addressed using the
technology
bull The incentives and disincentives for adopting new technology must be equita-
ble not exploitative and aligned with effective use of the technology
bull Disparity-driven technology gaps should be explicitly recognized To the extent
possible provisions should be made to address the digital divide
Summary 9
Summary of Recommendations
The guidance document makes a number of recommendations related to (1) supporting effective and informed adoption of DCTT (2) designing flexible technologies to maximize public health utility while respecting other values (3) establishing meaningful processes for user disclosure and authorizationconsent (4) promoting equity and fairness in application of DCTT and (5) instituting transparent governance and oversight Here we provide a summary of recommendations
Supporting Effective and Informed Adoption
bull Those who authorize use of DCTT within a particular jurisdic-tion or institution should continuously and systematically monitor the technologyrsquos performance in that context This should include monitoring for effectiveness and benefit monitoring for harms and monitoring for the fair distribution of both benefits and harms They should also monitor evidence that is being generated in other contexts about their selected technological solution and about other competing technologies
bull Data collected through DCTT should be made available to public health professionals and to researchers in de-identified form to support population-level epidemiologic analysis
bull Data should be available to users that would permit them to further investigate their personal risk with public health officials or other health workers to add a layer of protection against unnecessary quarantine
bull Technologies or apps may produce some false negatives or false positives but they should be accurate enough that public health authorities feel confident that they support and donrsquot detract from contact tracing efforts
bull Trusted leaders should be enlisted to communicate effectively with the public about DCTT and encourage its use should the technol-ogy demonstrate some potential The limits of knowledge regard-ing effectiveness should also be explained along with what will be done to improve technological capabilities as understanding evolves
10 Digital Contact Tracing for Pandemic Response
bull Incentives can be a useful complement to encouragements how-ever any incentives for users to install and use DCTT must be equitable should not be coercive and should align with effective use of the technology
bull DCTT use should not be mandated at this time given uncertainty about potential harms and benefits Additional technology user and real-world testing is needed
Designing Flexible Technology to Maximize
Public Health Utility While Respecting Other Values
bull Technology companies should not alone control the terms condi-tions or capabilities of DCTT nor should they presume to know what is acceptable to members of the public
bull A ldquovalues in designrdquo approach to development of DCTT should be adopted (Flanagan Howe and Nissenbaum 2008 Knobel and Bowker 2011) Robust public and user engagement activities should be pursued to identify and incorporate to the extent pos-sible a range of values into the design of the technology These values may include privacy but also autonomy efficiency equity or others Technology design should reflect an appropriate balance and prioritization of identified values
bull Technology design should not be static but should be capable of evolving depending upon local conditions new evidence and changing preferences and priorities
bull DCTT should be designed to have a base set of features that pro-tect privacy with layers of additional capabilities that users may choose to activate An initial default should be that user location data are not shared but users should be provided with easy mech-anisms and prompts to allow for opting-in to this capability with encouragement to the public if it is shown to be critical to achieving public health goals
Establishing Meaningful Processes for
User Disclosure and Authorization (Consent)
bull A clear and concise module consisting of basic disclosure and vol-untary authorization should be developed to accompany DCTT
Summary 11
This module should not take the form of ldquoclickwraprdquo terms of service or end-user agreements but rather provide only essential information necessary for an individual to make a decision More detailed disclosures (such as FAQs in plain language) should be made easily accessible to those who wish to learn more with no hidden surprises
bull An opt-in approach to authorization should be instituted to ac-company initial DCTT rollout The feasibility and value of opt-out approaches should continue to be evaluated informed by what is technologically possible what local assessments of benefits and harms of the technology reveal over time and our evolving un-derstanding of the degree to which an opt-out approach is likely to increase or decrease utilization among different populations Opt-out approaches should not be precluded
Promoting Equity and Fairness in Application of DCTT
bull States localities and institutions that recommend widespread use of DCTT should provide technology (eg mobile phones Blue-tooth devices) and free data packages to those who desire but lack access to these devices
bull If there are lower rates of adoption of DCTT systems in some iden-tifiable communities public health authorities should find ways to compensate For example directing more non-DCTT resources and efforts toward those communities to meet specific needs that are elsewhere being supported by technology
bull If maps are generated based on DCTT to provide the public with the locations that COV+ individuals have visited steps must be taken to minimize the stigma and potential financial losses that could result from a location being identified as a hotspot
Instituting Transparent Governance and Oversight
bull Digital surveillance oversight committees should be established expeditiously with diverse and qualified membership to provide ethical and regulatory review prior to and concurrent with wide-spread use of a DCTT system
bull Understandable and publicly accessible rules must guide the col-
12 Digital Contact Tracing for Pandemic Response
lection access control use storage and combination of data by government authorities public and private institutions and other parties such as public health researchers
bull Only those data that are necessary and relevant for the public health response to COVID-19 should be collected and used
bull Identifiable data should be kept only for the period of time needed for the public health response to COVID-19
bull Identifiable data collected as part of this response should not be shared with anyone other than the relevant public health authorities without additional specific informed consent of individual users
bull Before a government or institution adopts a digital contact tracing program they should state the conditions under which the digital contact tracing program will be terminated
bull Future use of DCTT to advance public health or other efforts (eg use in seasonal flu surveillance) would require independent justification DCTT designed for public health use should not be used by law or immigration enforcement
bull The principles offered in this guidance document apply both during and following the COVID-19 pandemic
Legislative Recommendations
bull The United States Congress should enact new legislation specif-ically tailored to facilitate the use of DCTT as part of the public health response to COVID-19 while also protecting user privacy and ensuring data security
bull Congress should require DCTT developers to disclose to users in clear language the nature of the information that would be col-lected how it would be collected how it would be stored and for what purposes it may be used
bull While the rollout of DCTT should initially employ an opt-in au-thorization approach the feasibility acceptability and value of opt-out approaches should continue to be evaluated As such opt-out approaches to consent should not be precluded by legislation
Summary 13
bull Congress should prohibit the commercial use of data collected for COVID-19 response by DCTT
bull Congress should prohibit discrimination on the basis of data col-lected by DCTT
bull If Congress is unable to enact suitable legislation state legislatures should work toward enacting similar laws for their jurisdictions A ldquomodelrdquo state law should be rapidly developed to facilitate na-tionwide adoption of an appropriate law and uniformity of legal requirements
Summary of Analysis
Supporting Effective and Informed Adoption
The COVID-19 pandemic and the physical distancing efforts imple-mented to slow the rate of transmission have caused severe harm to indi-viduals communities and our society To protect the public good going forward we need a robust public health response that reduces the spread of SARS-CoV-2 and does so in a way that allows economic recovery to occur and to be sustained We also need to design and manage this public health response so as to minimize harms to individuals and society to distribute benefits and burdens equitably across the population and to avoid misuses of the technologies and the data they collect
To reduce the spread of SARS-CoV-2 chains of transmission need to be broken To do this people who have been exposed to SARS-CoV-2 or potentially exposed need to be identified as comprehensively and as quickly as possible so they can quarantine themselves and avoid infecting others This is the job of manual contact tracing by public health authori-ties in which people infected or presumptively infected with SARS-CoV-2 are interviewed and asked about their movements and interactions in-cluding where they work and shop how they travel with whom theyrsquove had contact and the nature of that contact (eg where the contact took place) Their contacts are then interviewed and potentially asked to quar-antine seek testing and take other protective measures if the contact is sufficiently high risk
14 Digital Contact Tracing for Pandemic Response
The hope is that DCTT can augment traditional contact tracing ef-forts either by working alongside and independently of manual contact tracing or by being integrated into manual contact tracing efforts in a way that makes these efforts faster more thorough and more efficient
Data suggest that a substantial proportion of transmissionsmdashper-haps as high as 50mdashoccur between individuals who are not symptom-atic and that transmission may occur as early as 3 days before onset of symptoms (WHO 2020) Because asymptomatic spread of SARS-CoV-2 appears to be a significant source of infection we need to identify po-tentially infected people before they show symptoms thus speed is of the essence This is one benefit of using DCTT potential contacts can be identified instantaneously notified quickly and asked to quarantine as soon as possible
Another benefit is identifying contacts who manual contact tracing methods may miss either because COV+ people do not remember all the places theyrsquove been or cannot identify all the people theyrsquove had contact with This is especially relevant given the long period of infectivity of SARS-CoV-2 which begins before people are symptomatic and aware they are infected (Ferretti et al 2020) If DCTT were designed to have optional location-monitoring capabilities this critical challenge could be mitigated even further For example location data might reveal that a COV+ person was at a restaurant at an exact time and date which could be followed up by contact tracers who could alert the public or use other measures to reach those who were also present in the restaurant at the same time In other disease contexts geolocation data have demonstrated some potential to support epidemiology and disease surveillance (see Fur-lanello et al 2002 Dredze et al 2013 Eckhoff and Tatem 2015 Fraccaro et al 2019) with technical cautions regarding accuracy and the like (Beu-kenhorst et al 2017)
One role for DCTT is to work alongside manual contact tracing but independently of it Individuals would download proximity tracing or exposure notification apps use them receive alerts if theyrsquove had a poten-tial contact with another user who is COV+ or presumptively COV+ and voluntarily self-quarantine without having contact with public health au-thorities or giving them data that feeds into public health contact tracing efforts It is possible that this would help to break chains of transmission and reduce the spread of SARS-CoV-2 though at this point these benefits
Summary 15
are speculative It is also possible that such exposure notifications will result in high rates of false positives
Another possible role is for DCTT to be integrated into manual con-tact tracing efforts When potential contacts are identified by DCTT they are connected to public health authorities who can then follow up with them There are different forms this could take and different kinds and amounts of data about contacts public health authorities could receive from DCTT On one end of the spectrum of reporting public health authorities would not receive individualsrsquo names or contact information only anonymous data The fullest version of reporting would securely send to public health authorities the names contact information such as address and phone number and other data about contacts that DCTT collected including data about their location and movement history
It is uncertain whether providing public health authorities with vol-umes of information on cases and contacts from DCTT will be useful in practice As mentioned above providing public health authorities with location data on cases and contacts collected by DCTT may help con-tact tracers to find and notify additional contacts However at present providing public health authorities with large amounts of data will be useful only if there is sufficient capacity to follow up on these data In addition there is a risk of low-quality data from DCTT flooding the system leading to investigation of false case contacts identified by DCTT and distracting from other important efforts Whether and to what extent data from DCTT will benefit contact tracing efforts is unknown pointing again to the importance of continuously collecting high-quality evidence about DCTT
Designing Flexible Technology to Maximize
Public Health Utility While Respecting Other Values
Use of DCTT is essentially an experiment as we have insufficient infor-mation about the performance of different DCTT and their efficacy In the face of this uncertainty how should DCTT be designed and how should its use be managed
Many efforts to advance DCTT in the United States and elsewhere have emphasized the importance of ldquoprivacy by designrdquo that is building privacy and security protections into the design of technology rather than counting on responsible use alone (Cavoukian 2010) As noted above
16 Digital Contact Tracing for Pandemic Response
some major technology companies have signaled this position through de-velopment of PPPT systems that embed features such as decentralization de-identified information user anonymity bans on collection of location data and minimal reliance on or integration of public health authorities or other government actors Many of these features have also been em-braced early by advocacy organizations (Crocker Opsahl and Cyphers 2020 Electronic Privacy Information Center 2020 Kahn Gilmor 2020) and in an open letter (ldquoJoint Statement on Contact Tracingrdquo 2020) from nearly 300 researchers These same actors have emphasized that use of DCTT should be fully voluntary
Although privacy is a key value individuals and communities may also value efficiency equity liberty autonomy economic well-being com-panionship patriotism or solidarity among other values People may accept more significant encroachments on privacy now if this ultimately results in realizing other values (such as companionship) that are of equal or greater importance to those individuals Rather than centering pri-vacy alone in design a different orientation is needed at this moment that of ldquovalues in designrdquo which incorporates a broader range of values into technology (Flanagan Howe and Nissenbaum 2008 Knobel and Bowker 2011) For example some users might wish to express autonomy solidarity or patriotism through DCTT by sharing their location history with public health professionals in order to advance the public health re-sponse increase system efficiencies (eg by contributing information that can lead to better data processing) and reduce the burden on essential workers At the same time there is value in further advancing autonomy by designing technology to allow individuals some control over what data about them are collected and shared
DCTT should be designed to have a base set of features that protect privacy and strive for interoperability but also should include other op-tional capabilities This could be achieved by designing DCTT to have a default that can be modified for example an initial setting could be that usersrsquo location data are not shared with public health authorities but us-ers may opt-in to this feature Such an opt-in approach is likely consistent with existing federal privacy laws
Designing DCTT this way gives users the flexibility to decide how to use the technology and how to engage with public health authori-ties consistent with their values and trade-offs they are willing to make This flexibility could also allow for more real-world evaluation of how
Summary 17
different users experience different features of DCTT in different loca-tions Technology design should not be static but it should be capable of evolving depending upon local conditions new evidence and changing preferences and priorities
DCTT developers must comply with a number of federal privacy laws These privacy laws generally permit the collection storage and use of personal information so long as the user provides meaningful consent Privacy law in the United States is generally sector-specific and limited in scope resulting in a patchwork of protections that differ significantly depending on the entity that collects the data and the type of data col-lected Given the complexity of existing federal privacy law and the need to further strengthen public trust in DCTT it would be beneficial for Congress to enact new privacy legislation that is specifically tailored to the use of DCTT in response to COVID-19 Such COVID-specific legisla-tion should be sensitive to the full range of values and recommendations described above
In short designing ldquomiddle-groundrdquo DCTT for flexible use may pro-vide the most adaptable and thus most robust public health responsemdashrespecting privacy and individual autonomy by allowing users to use DCTT in ways that express their own values
Public Acceptance of DCTT
While some groups have maintained that only PPPT-like minimal systems will be widely adopted because only they will earn and maintain public trust (Simpson and Conner 2020) there is insufficient evidence that pub-lic trust would be threatened by a DCTT system that has the capacity to securely collect location data integrate public health authorities and en-able voluntary sharing of certain user data (eg location data) with those authorities More research including through deliberative engagement sessions is needed to better understand how differences in the features and functionality of DCTT (such as optional sharing of geolocation data) influence trust and peoplersquos willingness to use DCTT Technology com-panies should not alone control the terms conditions and capabilities of DCTT nor should they presume to know what is acceptable to members of the public
Significant concerns have also been expressed by privacy advocates (Guariglia 2020) and in the popular press (Giglio 2020) about ldquosurveil-lance creeprdquomdashthat is a belief that state or corporate actors will use new
18 Digital Contact Tracing for Pandemic Response
surveillance technologies capacities and permissions well beyond the purposes for which they were initially justified to the public and beyond the time when they are useful for the COVID-19 pandemic Surveillance creep is a serious concern and should be carefully guarded against how-ever the possibility of surveillance creep is not a sufficient reason to limit development of DCTT to minimal systems Instead protections should be put in place to ensure that only those data that are necessary and relevant for the public health purposes at hand are collected and used and data should be kept only for the period of time needed for those pub-lic health purposes For this reason we would support COVID-specific legislation that would impose strict limits on the use of DCTT data for nonndashpublic health purposes
Finally the use of DCTT during the current pandemic should not set a precedent for future public health use (eg use in seasonal flu surveil-lance efforts) Future use would require independent justification Further use of DCTT in other contexts (eg by law enforcement or immigration enforcement) is presumptively unethical
Encouraging Adoption of DCTT
Researchers have estimated perhaps conservatively that DCTT use by 80 of smartphone owners or 56 of the population overall will be needed to suppress the epidemic (Hinch et al 2020) These estimates also highlight that some decrease in transmission would be realizable even with lower rates of technology adoption
In the United States many advocates and researchers have argued that use of DCTT must be fully voluntary However experience from other countries suggests that when use of a digital contact tracing app is voluntary only a minority of the population will download it Instead of making use fully voluntary and initiated by users there are ways that DCTT could be put into use without usersrsquo voluntary choice For exam-ple use of an app could be mandated as a precondition for returning to work or school or even further to control entry into a facility or trans-portation (such as airplanes) through scanning of a QR code to demon-strate personal exposure levels (Gan and Culver 2020)
While these approaches are hard to imagine in the United States some have argued that mandatory use of DCTT could be ethical If man-dates increase adoption of DCTT and improve the public health response
Summary 19
this would reduce the likelihood of lockdowns which are harmful and a severe limitation of individual liberty applied on a mass scale On the other hand mandated use of DCTT systems may not be effective People may not adhere to the mandate by simply leaving their phone at home Perhaps more important should the technology not deliver the hoped-for benefits having mandated the use of an unproven technology could result in a loss of public trust in the technology in the entity instituting the mandate and in the larger public health response potentially lead-ing to noncompliance with public health recommendations more broadly (Bernstein et al 2019)
Any decision maker considering mandatory use including govern-ment officials institutional leaders and employers must convincingly address a number of considerations Particularly important is the need to identify reliable evidence that the DCTT would be effective and to ensure that the burdens and benefits of use are equitable and justifiable At this time mandated use of DCTT by states or institutions is not jus-tifiable given uncertainty about potential harms and benefits Additional technology user and real-world testing is needed before mandatory use should be considered
As with any public health effort the amount of evidence that must be offered to illustrate that the intervention or program can achieve its aims and the degree to which people should be able to exercise choice in their participation should be in proportion to the anticipated bur-dens of the intervention or program For example the permissibility of mandating use of DCTT by the public depends on factors such as the sensitivity of the data that are collected the extent to which public health is integrated within the DCTT system and what actions are taken in response to confirmed virus exposure or being identified as COV+ (eg forced quarantine) The more burdens that are placed on individualsmdashfor example whether people are ordered into quarantine if they have been exposed to the virus or if there are limited social supports for those in quarantinemdashthe greater the demand should be on the performance of the DCTT system
Perhaps the most effective way to generate widespread US adoption of DCTT will be to offer incentives for its use in other contexts gener-ally speaking small incentives have been shown to lead to an increase in desired outcome (Singer and Ye 2013 Lee et al 2014) Given the impor-
20 Digital Contact Tracing for Pandemic Response
tance of widespread use modest incentives ought to be considered if and when there is sufficient evidence of the utility of DCTT so long as those incentives are not mandates in disguise Another ldquofirst linerdquo approach to increasing use of DCTT is for trusted community leaders public figures health care professionals and other respected individuals to communicate with the public and their communities about DCTT and to encourage its use through public engagement campaigns if and when the technology demonstrates sufficient potential
Establishing Meaningful Processes for
User Disclosure and Authorization (Consent)
Any effort to roll out DCTT should ensure that users have a meaning-ful opportunity to review and understand information about the specific technology and its uses Moreover given the importance of public trust and the current crisis of public trust in governments and technology com-panies handling private digital information there is a strong ethics argu-ment for requiring consent from individual users We recommend a care-fully crafted version of what is sometimes called simple consent which consists of basic disclosure and voluntary agreement or authorization This disclosure should include information about the purposes of the technology the userrsquos options for collecting and sharing data purposes for which data can be used and any known risks among other informa-tion This information should be presented in an accessible format on any DCTT app and more detailed disclosures should be readily accessible for those who wish to review them
Through an opt-in mechanism such as clicking a button to signal agreement users should be able to indicate their intention to use a DCTT The opt-in approach is consistent with mechanisms for agreement to use other downloaded applications An opt-in approach should be part of the initial introduction of DCTT given the novelty of the technology and its uses and the need to build trust and confidence in the system Successes of opt-out approaches in other areas suggest that the feasibility and value of an opt-out approach to DCTT should be carefully evaluated particularly in conjunction with assessment of whether public health goals are being met (Rithalia et al 2009) Such assessments should be informed by what is technologically possible by local data regarding benefits and harms of the technology and by evolving understanding of the degree to which
Summary 21
an opt-out approach is likely to increase or decrease utilization among different populations
Promoting Equity and Fairness in Application of DCTT
Digital contact tracing technology should be designed and used in ways that as far as possible promote an equitable distribution of benefits and burdens DCTT should be deployed in a manner that does not propagate preexisting patterns of unfair disadvantage or distribute harms and risks unfairly throughout the population It is well known that some commu-nities have lower rates of technology and data access and therefore may benefit less from use of DCTT unless steps are taken to address these digital disparities Additionally should use of DCTT be made a require-ment for entry into a workplace into a school or onto transportation then those who currently do not possess the required technology must not be unfairly burdened through lack of access In order to mitigate this states localities and institutions that recommend widespread use of DCTT should provide technology (eg mobile phones Bluetooth de-vices) and free data packages to members of the community who desire but lack access to these devices
Some populations may also experience greater harm and greater fear of harm from having their data collected For example some groups such as African Americans Hispanic Americans Muslim Americans and undocumented immigrants have more reasonable fear of their data being handed over to law or immigration enforcement and some groups have lower levels of trust in public health due to past injustices (Auxier et al 2019 CSM 2017 Pew Research Center 2017 Rodrigues et al 2018) This further substantiates the need to limit use of any data gathered by DCTT to its public health purpose
Instituting Transparent Governance and Oversight
DCTT must be developed with an eye toward both present and future implications We are rapidly gaining knowledge about SARS-CoV-2 and COVID-19 but still have essential gaps in our understanding In the United States public health responses including DCTT will generally be developed and coordinated by individual states regional consortia (Res-ton Sgueglia and Mossburg 2020) and associations Good governance in this context requires transparency and the creation of oversight bodies
22 Digital Contact Tracing for Pandemic Response
with the appropriate expertise and representation to allow nimble and effective responses while serving as trusted representatives
In order to address the range of ethics and governance concerns that relate to the design and use of DCTT we recommend that digital surveil-lance oversight committees be established perhaps at a state level and with a platform for national coordination These committees can provide ethics and regulatory review prior to and concurrent with widespread use of DCTT The committees should be composed of a diverse group of experts capable of evaluating the quality of a DCTT system locally including members of communities that experience higher rates of digital disparity
When assessing the design and use of digital contact tracing systems these committees (and the public more widely) should consider not only the risks and benefits accrued during the COVID-19 pandemic but also implications for the future How can we navigate safe use of these tech-nologies in a way that preserves public trust in them and enables the possibility of future beneficial use
As a start it should be emphasized that the principles offered in this and other guidance documents do not apply only during the pandemic Future efforts to advance DCTT capabilities during quieter times should make every effort to follow them
23
Introduction
Public health professionals around the world are working tirelessly to respond to the COVID-19 pandemic using tried-and-tested public health methods for infectious disease surveillance and control These traditional methods are essential to the global COVID-19 response To complement these actions and potentially augment the speed and efficacy of the public health workforce digital technologies are being harnessed Given the scale of the pandemic significant efforts are being undertaken to develop and leverage public-facing and health-system-supportive tech-nology solutions including smartphone apps and other digital tools that may aid public health surveillance and contact tracing
Digital contact tracing technology and closely related digital health products (hereafter DCTT) have been used in several countries as part of broader disease surveillance and containment strategies Globally many digital COVID-19 contact tracing strategies have already emerged in response to the pandemic This is not surprising given the ubiquity of mobile phones and other digital devices around the world (ldquoMobile Cellular Subscriptions (per 100 People)rdquo 2018) experiences developed during prior outbreaks and pandemics and the pre-COVID-19 momen-tum behind using digital technologies to support individual and health system capabilities (WHO 2017 Mathews et al 2019 Aiello Renson and Zivich 2020 Mahmood et al 2020) In the United States DCTT has been proposed as an integral part of some plans to ldquoreopenrdquo the country (Allen et al 2020 Hart et al 2020 Simpson and Conner 2020) It is al-most certain that these and related technologies will become part of not only the COVID-19 response but also the larger toolbox for future public health communicable disease prevention and control
24 Digital Contact Tracing for Pandemic Response
While novel public health surveillance technologies such as DCTT have theoretical promise their effectiveness is unclear These technologies also raise important ethical legal and governance challenges that require comprehensive analysis in order to support decision-making regarding their appropriate use A number of frameworks recommendations and analyses have emerged recently in an effort to chart potentially ldquosaferdquo pathways for use of public health disease surveillance technology Many in the United States such as the Electronic Frontier Foundation Elec-tronic Privacy Information Center American Civil Liberties Union and the Center for American Progress are proposing that digital public health surveillance technologies must embrace strict data privacy protections decentralized data storage a high degree of anonymity and voluntary adoption (Crocker Opsahl and Cyphers 2020 Electronic Privacy Infor-mation Center 2020 Kahn Gilmor 2020 Simpson and Conner 2020) Others have argued that technologies that seek to enhance public health response during a pandemic should more closely align with the needs of public health professionals and the evidence-based procedures they follow stating that interests in serving the publicrsquos health ought to weigh more heavily in the necessary balancing of stakeholder interests (de Jong et al 2019 Watson et al 2020) This view is in part based on a recogni-tion that during countless other outbreaks the public has benefited from traditional disease surveillance and contact tracing which are heavily re-liant on centralized data storage and when necessary the collection of identifiable information These traditional approaches are governed by ethics principles (PHLS 2002) ethics guidelines (WHO 2017) and laws (ASTHO 2012) and digital technologies represent a new tool to support them
While debates and recommendations about appropriate design and use of DCTT have focused intensely on minimizing important data-related risks a wider lens is needed to fully appreciate the many additional criti-cal questions that need attention This report begins to grapple with these questions which are critical to address in order to guide responsible use of DCTT Given the complexity of the terrain as a first step toward estab-lishing a foundation for responsible decision-making regarding potential use of DCTT we offer a set of guiding principles (see box) These prin-ciples are meant to apply to DCTT as well as other digital technologies used in novel ways during pandemic response
Guiding Principles for the Use of Digital Public Health Technologies for Pandemic Response
Transparency and public engagement are essential to an inclusive digital public health response
bull Government public health and digital technology leaders must engage
effectively with the public and other stakeholders to communicate the util-
ity importance oversight and limitations of relevant digital technologies
including their implications for individualsrsquo privacy and civil liberties
bull Transparency at all levels is essential for maintaining public trust and
confidence
bull To the extent possible digital public health responses should reflect the
range of values that are important to individuals including advancing the
health and well-being of the community as a whole
bull Decision makers should recognize the sacrifices that some people may be
willing to make during a pandemic in order to advance public health goals
Acceptance by some of particular monitoring capabilities should not be read
as a willingness to extend these methods to other problems or uses
Digital public health responses must represent the least infringement of civil liberties necessary to accomplish the public health goals
bull If preferred digital public health strategies infringe on privacy and other civil
liberties the infringements must be sufficiently justified by the circum-
stances of the pandemic offset by ample anticipated public benefit and
considered relative to infringements associated with other possible strate-
gies such as mass physical distancing
bull Only those data that are necessary and relevant for the stated public health
purposes should be collected Identifiable data should be stored in a secure
manner and only for the period of time that the public health purposes
require
bull Adopted technologies should not be used in ways that subject communities
to discrimination or surveillance for nonndashpublic health reasons
bull Respect for individual autonomy requires that users are sufficiently informed
of the public health goals of the technology and the extent to which those
goals are being met
Use of digital public health technologies and data must be guided by best available evidence
bull Decisions to deploy digital public health technologies should be based on a
careful assessment of the uses and limitations of any proposed technology
taking into account the best available evidence
bull Those who deploy digital public health technologies should continuously and
systematically monitor their performance as well as any evidence that is being
generated in other contexts about the selected technological solution and about
other competing technologies
bull Unintended consequencesmdashincluding those that might impact public health
goals core values and interests of the public and unfair advantage or disadvan-
tagemdashshould be carefully monitored and addressed as necessary
Responsible use of digital public health technology requires meaningful governance and accountability
bull Systems of governance must be trustworthy and well informed They must be
reviewed and adjusted as circumstances and evidence change or as unintended
effects are identified
bull Trusted representatives who are capable of developing and implementing uni-
form and fair standards for adopting and utilizing underlying digital technology
must be identified
bull Understandable transparent and publicly accessible rules must guide the
collection access control use storage and combination of data by government
authorities public and private institutions and other parties such as public
health researchers
bull Oversight accountability and consequences for abuse or misuse of these data
must be explicit and enforceable
The deployment of digital public health technology must be rooted in a commitment to equity
bull Digital public health technologies should be deployed in a manner that does
not propagate preexisting patterns of unfair disadvantage or further distribute
harms and risks unfairly throughout the population
bull To the extent possible digital public health technologies should be designed to
rectify existing inequities
bull Oversight mechanisms must be in place to ensure that the improved public
health outcomes are equitable and to detect and correct any unforeseen resul-
tant injustices attributable to the technology or that can be addressed using the
technology
bull The incentives and disincentives for adopting new technology must be equita-
ble not exploitative and aligned with effective use of the technology
bull Disparity-driven technology gaps should be explicitly recognized To the extent
possible provisions should be made to address the digital divide
Introduction 27
In reflecting on these principles it becomes clear that if we wish not only to realize but to maximize the public good that might come from use of DCTT we must carefully define and responsibly align public health needs and capabilities with technological needs and capabilities We must understand that although technology may serve as a workforce multiplier it alone will not solve the public health challenges we face We must identify and address assumptions and misinformation about technologies and data use We must provide the means and opportu-nity for informed decision-making by the public and those who serve as our representatives Government officials public health leaders leaders of other institutions employers digital technology developers and the public all must be adequately informed and engaged in order to make the best decisions possible under the circumstances
ONE
29
Types of Information Collected through Contact Tracing
Data Collected from Infected Persons
Symptoms and Course of Illness
Information about COVID-19 patientsrsquo signs symptoms and course of illness is important to public health because it provides a basis for refining clinical case definitions and informing health care providers and the gen-eral public (CDC 2020c) This includes the specific signs and symptoms manifested by persons who are COV+ as well as the relative frequency and durations of different signs and symptoms This would also take into consideration those persons with no symptoms but who test positivemdashthose who are presymptomatic (develop symptoms later) those who are postsymptomatic (clinically recovered but still infectious) and those who never manifest illness at all
Typically contact tracing begins with a case in which a person has confirmation of infection by means of a diagnostic test However in some cases test results are not reported until several days later and individuals may be identified as ldquopresumptive positiverdquo cases until testing can be completed In these cases contact tracing efforts will need to be updated when test results are returned For example if a test comes back negative public health professionals will want to notify contacts that they no lon-ger need to quarantine
Public Health Perspective
30 Digital Contact Tracing for Pandemic Response
Movement and Contacts
In order to manage cases appropriately (identify and track the infected isolate the sick quarantine the exposed) public health officials need de-tails on each case (Resolve to Save Lives nd) First they need to know who and where the individual is That means personally identifiable infor-mation and contact information (address phone numbers email) It also means information about the nature intensity and duration of contact with individuals to whom they may have transmitted the disease This may include information about where the individual works and the kind of work they do (eg health care worker) how they travel (eg bus sub-way car) and where they shop or any other public venues they may have visited during a period of possible infectiousness (PIH 2020a) It may be helpful in certain circumstances for public health officials to ensure that suspected cases contacts or other high-risk individuals are following iso-lation and quarantine recommendations or orders
Contact tracing involves identifying all individuals who have had sig-nificant exposure to confirmed or probable cases during the time prior to and after the onset of symptoms both of which are times when the case is thought to be infectious (Africa CDC 2020) Contacts could be those who are caring for COVID-19 patients especially if they lacked proper PPE and those who had close interaction with the COV+ person over a sustained period of time particularly in enclosed spaces (PIH 2020a) For COVID-19 contacts are identified by asking a person with a confirmed or probable case about people they may have been within 6 feet of for 15 minutes or more starting from 48 hours before the onset of symptoms and lasting until the person is isolated (CDC 2020b)
Data Collected from Contacts of Infected Persons
Contact Details
In addition to the data collected from individuals with COVID-19 con-tact tracers will collect data from potentially exposed individuals (con-tacts) Information about the nature intensity and duration of contact with an infected person may be collected for a contact if information about the case is known to the contact These details can help a contact tracer more accurately determine whether the contact is at high or low risk for SARS-CoV-2 transmission and help determine whether a con-tact should quarantine for 14 days (the upper bound of the SARS-CoV-2
Public Health Perspective 31
incubation period) In addition public health professionals may gather contactsrsquo demographic information and other personal data to contribute to population-level disease surveillance and situational awareness about an epidemic (CDC 2005) However the information needed at baseline is only a personrsquos name and contact information
Symptoms (If They Develop) and Course of Illness as well as Information about Close Contacts
If a contact develops COVID-19 symptoms while in quarantine andor tests positive for the virus public health will then collect the data required for a COVID-19 case This includes collecting information on the con-tacts that a person may have had (if any) in the days immediately before and during the course of their infection
How Contact Tracing Information Informs Public Health Action
To reduce disease burden and help make ldquoreopeningrdquo safer during the COVID-19 pandemic the United States and other countries will need to identify gather information about and safely isolate cases and quar-antine their contacts to reduce community transmission (Watson et al 2020) Gathering information about possible cases and their contacts en-ables public health to break chains of transmission
Contact tracing involves stages (CDC 2020a) including
1 identifying an infected person as a COVID-19 case
2 identifying the close contacts of that case (Africa CDC 2020)
3 getting in touch with contacts
4 asking contacts to quarantine at home for 14 days
5 assessing contacts for possible symptoms and
6 following up with COV+ persons and their contacts to identify new or worsening symptoms and connect them with medical care if needed
Contact tracers also play an important role in providing resources for COV+ persons who are in home isolation and their contacts who are
32 Digital Contact Tracing for Pandemic Response
in home quarantine Knowing who and where cases and contacts are can enable provision of supplies such as digital thermometers or masks Effective contact tracing that enables isolated cases and quarantined con-tacts to remain at home also requires providing a range of social sup-port services or ldquocare packagesrdquo from delivering food and medicines to trash pickup Furthermore vulnerable individuals who are homeless or otherwise unable to sufficiently isolate or quarantine in their current living conditions may need to have alternative housing arranged to safely remain separated from others (CDC 2020b)
Finally contact tracers explain what quarantined contacts should do if they begin to develop symptoms consistent with COVID-19 (Africa CDC 2020) Depending on the context contact tracers may engage in active monitoring by regularly communicating with contacts about their health status through phone text message or possibly mobile applica-tions In rare cases public health can make quarantine mandatory and may monitor a quarantined individual to ensure that they do not break quarantine Contact tracers may also facilitate access to health care by providing telemedicine resources or other information and support for accessing medical care
Characteristics That Make Data Useful to Public Health for Reducing Disease Transmission
Data Access
If digital contact tracing technology and closely related digital health products (together DCTT) are intended to support the public health ac-tions described above and directly amplify public health capacity to con-duct case identification and contact tracing then data collected through DCTT must be accessible to public health authorities Identifying infor-mation and location data for cases and contacts of cases are necessary for public health use so that contact tracers can do their work to uncover ongoing transmission and enable isolation and quarantine These data should also be durable meaning that public health can return to the data in order to interact with and support cases and contacts These data can also be useful at a population level if de-identified and aggregated by illuminating trends in community transmission and providing support for decisions about resource allocation
Public Health Perspective 33
Data Format
Data should be provided to public health authorities in a usable format that is compatible with public health systems and that has the granularity and specificity of personal information that is needed for use in contact tracing Without personal identifiers the data cannot be used by public health workers to undertake contact tracing Data should also contain information about the nature of a contact including the proximity of the contact and number of minutes that the person was in contact with an infected individual Location data can also help public health author-ities to conduct contact tracing particularly when contact occurred in a crowded area and involved people who donrsquot know one another Loca-tion data from a case can help public health professionals identify con-tacts even when those contacts themselves are not using a contact tracing app because the data shows contact tracers where to look for additional contacts
Data Accuracy
Data that identifies individuals as having sustained contact with a case must be as accurate as possible If criteria for being considered a contact are too restrictive it may result in missed contacts and sustained chains of disease transmission If criteria are too broad it may result in unneces-sary restriction of movement which could have significant personal and economic consequences
Timeliness of Data
Data from cases and contacts must be timely in order to enable case-based management that will help reduce community transmission For contact tracing to be effective infected individuals need to be isolated and their contacts identified and quarantined as quickly as possible Testing for SARS-CoV-2 can take time sometimes many days for a test result Especially because SARS-CoV-2 is transmissible during the pre-symptomatic period data on symptomatic individuals should be made available to public health officials even before a positive test is returned in order to enable identification and quarantine of contacts right away If this information is delayed until a test result is received it may be too late to identify and quarantine contacts because contacts (if infected) will already be contagious and may have spread the virus to others
34 Digital Contact Tracing for Pandemic Response
VolumeAvailability of Data
The more that individuals opt to share their information to support con-tact tracing the more effective contact tracing will be in breaking chains of viral transmission and controlling epidemics of COVID-19 The exact proportion of cases and contacts that need to be identified in order to avoid large surges of cases which overwhelm health care systems is un-certain but the goal is to identify all infected cases and all close contacts of each case (PIH 2020b)
Recommendations
bull Technologies or apps with the goal of enhancing public health ca-pacity to identify cases and trace contacts in order to control the spread of SARS-CoV-2 should be designed to match functionality with that goal
bull Technologies or apps may produce some false negatives or false positives but they should be accurate enough that public health authorities feel confident that they support and donrsquot detract from contact tracing efforts
bull DCTT approaches for public health should be designed to facilitate the following
deg identifying contacts including those who may not be easily found otherwise
deg finding and notifying contacts rapidly before they develop symptoms if infected
deg analyzing the nature of contact to determine whether contact is high medium or low risk and to support decisions about whether a contact should quarantine and
deg following up with cases and contacts so that public health can provide resources to support isolation and quarantine
bull Data collected through DCTT should be made available to public health professionals and to researchers in de-identified form to sup-port population-level epidemiologic analysis
T WO
35
Characteristics of SARS-CoV-2 Relevant to Candidate Digital Solutions
The SARS-CoV-2 virus has some unique transmission characteristics and clinical manifestations that can help guide use of digital contact trac-ing solutions Individuals infected with this virus may or may not show symptoms or may show a range of different and sometimes nonspecific symptoms Estimates regarding the percentage of individuals who are in-fected but never develop symptoms is highly uncertain ranging from 5 to 50 (Heneghan Brassey and Jefferson 2020) Data suggest that a substantial proportion of transmissionsmdashperhaps as high as 50mdashoccur between individuals who are not symptomatic and that transmissibility may extend out as long as 3 days before the onset of symptoms (WHO 2020)
The complexity of asymptomatic and presymptomatic transmission makes it more difficult to identify all cases of COVID-19 It also means that manual contact tracing is less effective because people are unlikely to remember all of their contacts during the long period of infectivity (Ferretti et al 2020) however it does not negate the need for contact tracing Identifying symptomatic cases will still help greatly with slowing the spread because their contacts can be asked to quarantine to prevent them from spreading the virus if they are indeed infected This means that whether they become symptomatic or not contacts will be quarantined and the chain of transmission will be broken If contact tracing can be implemented on a large enough scale perhaps with support from DCTT
Digital Technology and Contact Tracing
36 Digital Contact Tracing for Pandemic Response
eventually the virus could be managed at much lower levels of community transmission and large epidemics of unrecognized spread will not occur
The transmissibility of the virus when a person has no symptoms further suggests that effective solutions may require multimodal inter-ventions combining contact tracing with frequent rapid and ubiquitous testing and continued social distancing to varying extents (Cheng et al 2020)
Because of presymptomatic spread contact tracing efforts and dig-ital solutions to augment those efforts should support identification of contacts a person had 2 days before their symptoms and at least 3 days after the resolution of those symptoms (if the person continued to have contacts through that time period) (CDC 2020d) Additionally public health messages delivered by these technologies should urge contacts to quarantine for the full 14-day incubation period
Previously Existing Contact Tracing Technologies
Prior to this pandemic health agencies in high- medium- and low-income countries had begun to develop and use digital tools to augment the man-agement of infectious diseases including sexually transmitted infections (HIV chlamydia gonorrhea) and high-consequence epidemics (Ebola) (Danquah et al 2019)) However these have been primarily used to facil-itate case interviews partner notification (in the case of STIs) and record keeping as opposed to fully digitizing or automating the contact tracing process
It has been recently suggested that digital contact tracing could con-tribute to the management of the ongoing COVID-19 pandemic and the experiences of containing SARS-CoV-2 in countries such as China Sin-gapore and South Korea provide noteworthy examples However un-dertaking this case-based intervention on the scale required to achieve pandemic control is a novelty in the history of public health Although technological development is proceeding rapidly several foundational is-sues have yet to be resolved including functionality connectivity to pub-lic health authorities and informatics systems usability by disease inter-vention specialists (DIS also referred to as contact tracers) and sufficient protection of personally identifiable information among others
Digital Technology and Contact Tracing 37
Introduction of Novel Digital Contact Tracing Technologies
Digital contact tracing technologies and platforms have recently been in-troduced and the CDC has published preliminary criteria for evaluating these tools (CDC 2020e) It can be helpful to consider three broad ap-proaches along the spectrum of potential methods of digital contact trac-ing a maximal approach (typified by the South Korean governmentrsquos cen-tralized data collection (M S Kim 2020)) a minimal approach (typified by the AppleGoogle decentralized privacy-protecting proximity tracking (Apple and Google nd)) and a diverse middle ground that aims to aug-ment manual contact tracing with the collection of digital data Perhaps the most promising approach in this middle ground involves allowing us-ers to turn over both proximity data and GPS location data (ie cell-site location data) to public health authorities on a voluntary basis
Along with this ldquominimal to maximalrdquo spectrum in the design of dig-ital contact tracing technologies and systems there is another spectrum that concerns voluntary versus mandatory use of these technologies are individuals entirely free to use these technologies or not or should poli-cies incentivize or even mandate their use At one extreme South Korea (Republic of Korea) implemented a system (called Safe Korea) supported by the Ministry of the Interior and Safety that collects a variety of per-sonal data in a centralized database in order to enforce quarantine orders and track possible contacts (M S Kim 2020) Israel also implemented a centralized involuntary data collection system for tracking COVID-19 cases and alerting those who may have been exposed (Hendrix and Eg-lash 2020) In Poland health authorities have set up mandatory ldquocheck-insrdquo involving a GPS-waypoint capture and ldquoselfierdquo photographs sent to the monitoring agency to ensure that individuals are not breaking quar-antine (Hamilton 2020)
These centralized systems can be designed to incorporate data from a variety of sources The data collected include location data from mobile phones QR codes can also be scanned to track the use of public transit where GPS data may be inadequate (due to low resolution) to accurately distinguish the occupants of one vehicle from another The data collected from mobile phones can then be integrated with data from other sources such as facial-recognition cameras credit card transactions and social media
38 Digital Contact Tracing for Pandemic Response
At the other extreme of technology invasiveness for contact trac-ing isolation and quarantine many corporations and working groups (including the AppleGoogle collaboration) have developed privacy-pre-serving proximity tracking (PPPT) using Bluetooth Low Energy (BLE) ldquohandshakesrdquo that record close contact between mobile phone users In most architectures these proximity data are stored in the usersrsquo phones as anonymized ldquobeaconsrdquo that cannot be used to re-identify the users directly If a user with a PPPT app installed on their phone tests positive and enters test results into their app those who have been identified as having been in close proximity to them can be notified by the app This notification can be automatic or at the discretion of the person who is COV+ depending on the app design If notified a user who has been in contact with a COV+ individual would receive a push notification alert-ing them to possible exposure (which may be timestamped) but with no other identifying information
Because of its reliance on anonymized data PPPT on its own is dis-tinct from manual contact tracing In recognition of this fact some de-signers and researchers now use the more descriptive term ldquoexposure no-tificationrdquo Moreover the public health usefulness of PPPT is uncertain it is unclear how PPPT can best be used in tandem with manual contact tracing especially if the data it collects are inaccessible to or unusable by public health authorities It remains to be seen whether PPPT will provide significant benefit operating alongside but not integrated into manual contact tracing
Between these extremes there are a number of possible middle-ground approaches that aim to strike a balance among public health utility tech-nological feasibility and user privacy protections This middle ground divides into two rough categories centralized storage of de-identified data and decentralized storage of personally identifying data The United Kingdomrsquos NHSX is reportedly developing an app that would utilize BLE handshakes to collect anonymized proximity data which would then be stored on a centralized government-operated server
The most prevalent middle-ground approach in the United States context involves the collection and storage of personal datamdashincluding identifying information and location datamdashon the userrsquos phone This de-centralized but personally identifiable data can then be voluntarily shared with public health officials if the user tests positive for SARS-CoV-2 For
Digital Technology and Contact Tracing 39
example an MIT team has developed an app called Private Kit Safe Paths (MIT nd) that stores usersrsquo location data on their phone for 28 days If a user tests positive she can voluntarily upload her location data to a website that is accessible only to public health officials Officials can then analyze this personally identifiable data and subsequently broadcast re-dacted and de-identified data to other users Healthy users would have access to these redacted location data of COV+ users but their own data would not leave their phones (The developers plan to incorporate BLE proximity data once available) Along similar lines the North Dakota state government has rolled out an app that stores both location data and proximity data on a userrsquos phone which can be voluntarily released by the user to public health authorities if the user tests positive (NDDoH 2020) At a minimum the storage of user location data can function as a ldquomemory aidrdquo if the user tests positive But releasing the data to public health authorities may help them analyze the spread of COVID-19 and alert individuals or groups that have been in contact with persons who are COV+ An overview of various DCTT apps and platforms as well as features that are relevant to this analysis are provided in Table 1
Because DCTTs are so new very little is known about their actual utility to public health authorities for controlling this pandemic Although multiple countries that have had success in greatly reducing transmission of SARS-CoV-2 have included DCTT in their response these countries have employed multiple simultaneous approaches to controlling the vi-rus including manual contact tracing and it is difficult to disentangle what made those responses successful Preliminary impressions from Ice-land may suggest that DCTT at least in that context had a small impact on reducing transmission ldquoespecially compared with methods of manual contact tracing such as phone callsrdquo (Hadavas 2020) This is with the highest public download rate of any DCTT app thus far
DCTTs have the potential to be helpful but they also have the poten-tial to distract from other public health efforts including manual contact tracing Concerns about implementation of DCTT from the public health perspective include that data generated may not be useful to public health authorities either because they donrsquot include detailed data to aid contact tracers or because the data are inaccurate (Mills Rodrigo 2020) DCTT if not calibrated well could be overly inclusive and create many false positives This would be harmful to those individuals being notified and
Purp
ose
Tech
nolo
gies
Use
dD
ata
Stor
age
Part
icip
atio
n
Proximity-based exposure notification
Digital contact tracing (DCT)
Bluetooth LE
GPS
SMS
Centralized
Decentralized
Mandator (actually or functionally)
Voluntaryopt-in
Gov
ernm
ent a
cces
sA
pp N
ame
Dev
elop
er o
r Cou
ntry
Max
WeC
hat
Alip
ayCh
ina
D
ata
com
es fr
om g
over
nmen
t so
urce
s lo
catio
n da
ta s
ent t
o po
lice
Intervention Type
Middle Ground
Trac
e To
geth
erSi
ngap
ore
M
anda
tory
gov
ernm
ent
acce
ss if
pos
itive
NH
SXO
xfor
dO
xfor
d
Gov
ernm
ent m
aint
ains
dat
a
Nex
tTra
ceFr
ed H
utch
inso
n Ca
ncer
Re
sear
ch C
ente
r
Gov
ernm
ent m
aint
ains
dat
a
but n
o st
orag
e
COVI
D
Safe
Path
sM
IT
Volu
ntar
y up
load
by
user
s
who
test
pos
itive
Aar
ogya
Set
uIn
dia
A
nony
miz
ed a
ggre
gate
Care
19N
orth
Dak
ota
In
agg
rega
te o
ptio
nal i
f pos
itive
Minimal
Covi
dSaf
eU
niv
of W
ashi
ngto
n
Non
e
Covi
dWat
chU
niv
of S
tanf
ord
amp U
niv
of W
ater
loo
To
val
idat
e te
st re
sults
CoEp
iCo
Epi
O
pt-in
to s
hare
BT
and
sym
ptom
lo
g w
ith C
oEpi
ser
ver
itoG
erm
any
N
one
pos
itive
resu
lts to
ito
serv
er
TAB
LE 1
Ex
ampl
es o
f Dig
ital C
onta
ct T
raci
ng T
echn
olog
ies
to S
uppo
rt A
ctiv
e Pu
blic
Hea
lth S
urve
illan
ce a
nd R
elev
ant F
eatu
res
Digital Technology and Contact Tracing 41
asked to quarantine unnecessarily and it could result in large proportions of the population remaining at home at any one time Individuals living or working in congregate settings could receive frequent notifications that would result in their inability to leave quarantine for long periods of time Finally public health authorities could also become inundated by data from these technologies and not have sufficient approaches to manage or analyze the incoming information
Relevant Differences between Manual and Digital Contact Tracing
There are several noteworthy differences between manual contact tracing efforts and use of DCTT First there is a significant amount of evidence regarding the effectiveness of manual contact tracing which is lacking for DCTT Second manual contact tracers interact with individuals who are confirmed or suspected cases and contacts of cases but not other members of the general public DCTT intervention would affect all users regardless of circumstances (though some more than others) Third manual contact tracing occurs most often through human-to-human encounters with the opportunity to clarify misconceptions address worries and express sympathy and other important affects DCTT can certainly incorporate sharing of important information and potentially communicate some af-fect but it currently lacks a range of other human capabilities and char-acteristics Fourth there typically are fewer data intermediaries in manual contact tracing (fewer entities handling data) in DCTT a valid argument could be made that a wide range of technology developers (and perhaps mobile network operators) must remain connected to relevant data in order to continuously identify problems and improve functionality
It is because of these and other differences that DCTT has been pro-posed as a potential complement to rather than a replacement for man-ual contact tracing However over time it is possible that technology could develop to close gaps between some of these differences (if and as needed) and in parallel the goals of contact tracing and public health surveillance may evolve
Ethics of Designing and Using DCTT
43
THREE
43
Those developing DCTT and those considering its use should systemat-ically take into account and document alignment with the guiding princi-ples outlined in this report
When considering the ethics of DCTT key ethical questions con-cern the features that DCTT should have (eg should digital contact tracing apps collect usersrsquo location data) whether and how individualsrsquo data should be shared with public health authorities how ethically to encourage use of DCTT (eg under what circumstances would it be eth-ical to incentivize or mandate use of DCTT) what kind of supports and equity-promoting measures should accompany use of DCTT and how governance and oversight of DCTT should be structured
The sections that follow consider these questions one by one A key conclusion of this report is that these features of the design and use of DCTT are ethically interrelatedmdashreaching a determination regarding any one question requires careful consideration of them all Rather than reaching ldquoone size fits allrdquo conclusions about specific features of uses of DCTT decision makers should ethically assess DCTT systems holistically
Generally a public health measure is ethically justifiable if it strikes a reasonable balance between competing considerations and if it pro-vides sufficient public health benefit (or the prospect of benefit) to justify the burdens associated with it DCTT systems are ethically justifiable if they strike a reasonable balance between multiple ethical considerations including
bull enabling an effective and efficient public health response
bull protecting individual privacy and preventing harms to individuals
44 Digital Contact Tracing for Pandemic Response
including harms from sensitive data being revealed and from erro-neously being subjected to isolation or quarantine orders
bull allowing individuals to control what information about them is collected and revealed to whom including through appropriate dis-closure and authorization processes for data collection
bull promoting equitable distribution of benefits and burdens of DCTT
bull maintaining public trust in DCTT and in the COVID-19 public health response and
bull taking seriously the future implications of decisions that we make today
To illustrate a holistic assessment consider whether it is ethically jus-tifiable for an employer to mandate that employees use a DCTT as a condition of returning to work This will depend upon many features of the DCTT system what kind of data the DCTT collects (eg does it collect location data or just record proximity events) whether there is public health capacity to make good use of these data what the data are used for (eg will the employer ban an employee from the workplace on the basis of a DCTT-identified contact) what kind of social supports are available (eg is there paid leave for employees) what employeesrsquo attitudes are toward use of DCTT and whether mandating use is likely to have public health benefit among other factors These factors may vary from place to place and may change over the course of the pandemic Thus there is no ldquoone size fits allrdquo ethically optimal approach to DCTT
Justifying the Use of DCTT Systems
A foundational issue is why deploying any DCTT during a pandemic is justified given there are manual contact tracing capabilities that are well established while the performance and effectiveness of novel technolo-gies is less established The need to move quickly to minimize the spread of the virus poses challenges here as the data needed to fully make the case that these technologies substantially contribute to the public health response may not be available prior to widespread use The primary ar-gument for DCTT is that the capacity of manual contact tracing may be
Ethics of Designing and Using DCTT 45
exceeded and we may not be able to bolster the public health workforce rapidly and sufficiently enough to meet needs DCTT has the potential to quickly and exponentially expand the reach of contact tracing In ad-dition DCTT may allow more efficient identification and quarantine of potential contacts of COV+ people than manual contact tracing alone particularly given the high number of infections that have been spread by asymptomatic individuals
Nonetheless reasonable people disagree about the prudence of pur-suing DCTT especially given its limited performance history and poten-tial risks including diverting attention and resources from more effective interventions The limited attention and resources available during a pan-demic must be allocated efficiently and effectively
To justify potentially widespread use of technologies such as DCTT therefore a number of considerations must be addressed
bull whether the technology is designed to meet an important and unmet public health need
bull whether there is sufficient evidence or reason to suggest that the technology will be effective at serving its purpose
bull whether the outbreak is characterized by sufficiently severe morbid-ity and mortality and a high rate of disease transmission to warrant large-scale introduction of novel systems
bull whether there are other less autonomy-restricting or less risky al-ternatives to widespread use and
bull whether it is reasonably likely that a sufficient number of individ-uals will use the technology to achieve the intended public health benefit
Monitoring and Evaluating Technologies to Inform Policy and Practice
A number of public health ethics principles necessitate the ongoing mon-itoring and evaluation of DCTT systems First DCTT must be shown to perform reasonably well at achieving its stated goal reducing the spread of SARS-CoV-2 The effectiveness of DCTT programs should be illus-trated at a number of stages
46 Digital Contact Tracing for Pandemic Response
1 Robust initial technology testing is needed to publicly justify the widespread adoption of DCTT and avoid public failures which may hamper future uptake (eg Lovejoy 2020 Morse 2020) This typically includes alpha testing in virtual environments and beta testing in different community settings
2 If and when a DCTT is implemented on a wide scale it must be monitored on an ongoing basis to assess reach effectiveness func-tionality best practices and any harms
3 When approaching a previously identified stopping point for use of DCTT monitoring can help to identify when utilization is no longer needed
If at any of these points evidence clearly suggests harm (particularly in comparison to other methods that the public might find more acceptable) this evidence should provide a basis upon which to revisit strategies pri-orities and allocation of resources Attention should be given to foresee-able side effects that may dramatically influence the overall effectiveness of the program such as individuals carrying their smartphones around with them selectively so as to avoid particular undesired consequences of DCTT policies
Anonymized aggregate data including user feedback must be eval-uated to ensure that benefits and burdens are distributed fairly As noted earlier unintended burdens may include inequitable outcomes that may arise in a DCTT program for example resulting from uneven access to the required technology to participate disparate concerns about sur-veillance within some communities that might limit widespread use or discrimination that may result from being identified as COV+ due to the program or for communities that are termed ldquohotspotsrdquo based on maps of COV+ location data Additionally it is possible that some communi-ties might get higher rates of false positives because they are located in densely populated areas thus increasing the burden of self-isolation If any of these inequities are identified steps must be taken to mitigate them
Finally numerous actors should engage in the monitoring and eval-uation of DCTT systems Technology developers and public health re-searchers have a clear role in this process Technology developers should work with public health researchers to monitor accuracy precision func-
Ethics of Designing and Using DCTT 47
tionality confidence of estimates sources of error and the like Research-ers may also be able to contribute innovative methods to systematically and rapidly evaluate candidate technologies such as by deploying cluster randomized stepped wedge (Hemming et al 2015) or adaptive trial de-signs and techniques (eg response-adjusted randomization) (Pallmann et al 2018) These approaches were also proposed for use in research to assign candidate experimental treatments and vaccines during the 2014-15 Ebola outbreak (Berry et al 2016) When formal research activities are pursued ethics principles and legal requirements for the conduct of research should apply (eg The Belmont Report)
Furthermore any workplace or institution that incentivizes or man-dates use of DCTT has a responsibility to provide evidence that the in-tervention at minimum is not likely to cause harm and to monitor for unanticipated burdens In all cases it is vital that a trusted intermediary be involved in the evaluation of DCTT programs to limit perceptions of bias and ensure a legitimate basis for decision-making Nonsensitive aggregate DCTT analyses should be made available to the public so as to permit verification and inform continuing public debates about its useful-ness and necessity At an individual level data should be available to users that would permit them to further investigate their personal risk with public health officials or other health workers This is important not only to ensure their health and well-being but also to add a layer of protection against unnecessary quarantine
Recommendations
bull Reviews of DCTT systems must be conducted in part by an inde-pendent intermediary that has established the publicrsquos trust
bull Those who authorize use of DCTT within a particular jurisdic-tion or institution should continuously and systematically monitor the technologyrsquos performance in that context This should include monitoring for effectiveness and benefit monitoring for harms and monitoring for the fair distribution of both benefits and harms They should also monitor evidence that is being generated in other contexts about their selected technological solution and about other competing technologies
48 Digital Contact Tracing for Pandemic Response
bull Data should be available to users that would permit them to further investigate their personal risk with public health officials or other health workers to add a layer of protection against unnecessary quarantine
Public Trust and Public Attitudes
Researchers have estimated perhaps conservatively that DCTT use by 80 of smartphone ownersmdash56 of the populationmdashwill be needed to suppress the epidemic (Hinch et al 2020) These estimates also highlight that some decrease in transmission would be realizable even with lower rates of technology adoption As such in order to maximize impact it is essential to gain a thorough understanding of public perspectives on DCTT including which features and uses of the technology the pub-lic finds acceptable which kinds of DCTT the public would be most likely to use and which designs and uses of DCTT would maintain or jeopardize public confidence and trust There will be variation in public attitudes within and across societies and over time
With respect to what we currently know about public attitudes and trust in DCTT in the United States polling data suggest some potential support and also some divisions regarding willingness to use the technol-ogy Polls conducted by groups based at the University of Zurich (Hargit-tai et al 2020) and the University of Oxford (Altmann et al 2020) suggest that more than 60 of Americans would be willing to install such an app Both a Washington PostndashUniversity of Maryland poll (2020) and a Kaiser Family Foundation poll (Kirzinger et al 2020) show roughly half of the population would be willing to install the app Over half of the population (59) would be willing to share their COVID-19 positive test result with an app in order to anonymously share that information with their contacts (Washington PostndashUMD 2020) Only 29 of respondents to a March 12ndash27 Oliver Wyman Forum poll (Elliott et al 2020) said that they would be willing to share their location data Additionally Washing-ton PostndashUMD data and Pew data from 2019 suggest that approximately one in six Americans do not have a smartphone and thus cannot use the technology without intervention (Pew Research Center 2020)
People may be more willing however to download an app if it will
Ethics of Designing and Using DCTT 49
ease social distancing policies and allow for more economic and social activity Willingness to install a contact tracing app increased among re-spondents to the Kaiser Family Foundation poll from 50 to 66 when respondents were asked if they would be willing to do so to allow schools and businesses to reopen Additionally who develops or administers the app appears to matter Respondents to the Washington PostndashUMD poll indicated higher levels of trust that their anonymity would be preserved by public health agencies and universities than by tech companies or health insurance companies Further more respondents to the Oliver Wyman Forum poll were willing to share their health information with public health authorities (55) than the local government (35) their employer or school (33) or the federal government (27)
These data suggest that people will be more willing to use a contact tracing app when the potential benefits are clearly identified and valued such as lifting social distancing measures and they will be more willing to do so if the data are going to a public health agency rather than the federal government or a tech company Other factors that seem to be as-sociated with greater willingness to install a contact tracing app include younger age and the app source (Hargittai et al 2020) with a preference for apps distributed by public health agencies over others such as health insurers or public universities (Hargittai and Redmiles 2020) However all of this must be read with caution as public polling may not be repre-sentative of some populations or of widespread public attitudes Further these attitudes may shift over time and may be discordant with behaviors (Barth and de Jong 2017)
Deliberative public engagement efforts would be an appropriate means of filling in gaps in understanding about the acceptability of dif-ferent approaches (Fishkin and Laslett 2003 Cavalier 2011) In addi-tion including the public particularly in the earlier stages of planning a path to sustainable resolution to the pandemic could serve to help disseminate a nuanced understanding of what is at stake including the key challenges and trade-offs Aggregated public polling results are not sufficient as a proxy for careful analyses of the ethical challenges but they do provide a necessary input for these analyses Integrating lessons and outputs from public engagement into guidance and other products requires special attention and should be validated and enhanced through further engagement
50 Digital Contact Tracing for Pandemic Response
Recommendations
bull More research into public attitudes is needed In particular in-depth qualitative research should examine public attitudes about percep-tions of trust in DCTT among different communities which fea-tures of DCTT influence trust and the extent to which people are willing to provide different types of data through DCTT to help their community
bull States and localities that are considering adopting DCTT should engage with the public to increase their understanding of the ac-ceptability of DCTT design features and uses among diverse communities
Designing Flexible Technology to Maximize Public Health Utility While Respecting Other Values
Values in Design
Efforts to advance DCTT in the United States and elsewhere have empha-sized the importance of ldquoprivacy by designrdquo that is building privacy and security protections into the design of technology rather than counting on responsible use alone (Cavoukian 2010) As noted above some major technology companies have signaled this position through development of decentralized privacy-preserving proximity tracking (PPPT) systems These systems embed features such as decentralization anonymity of us-ers bans on collection of location data and minimal reliance on or inte-gration of public health authorities or other government actors Many of these features have also been embraced early by advocacy organizations (Crocker Opsahl and Cyphers 2020 Electronic Privacy Information Center 2020 Kahn Gilmor 2020) and in an open letter from nearly 300 researchers (ldquoJoint Statement on Contact Tracingrdquo 2020)
Privacy by design provides principles that incorporate one set of val-ues (privacy) into the design of DCTT Importantly the principles ac-knowledge the need to design privacy defaults into systems while main-taining the capacity of those systems to achieve their otherwise justifiable ends Put another way privacy by design ldquoembraces legitimate non-pri-vacy objectives and accommodates them in an innovative positive-sum mannerrdquo (Cavoukian 2010 p 4)
Ethics of Designing and Using DCTT 51
This stance simple in its statement is not easy to satisfy Given that ldquoobjectivesrdquo are themselves driven by values it begs for an articulation of additional values (aside from privacy) that individuals and groups within societymdashincluding many privacy advocatesmdashmay believe to be important For example at any moment in addition to valuing their own privacy individuals may value efficiency equity autonomy economic well-being companionship patriotism or solidarity Moreover the above stance necessitates an acknowledgment that peoplesrsquo value priorities often change when circumstances change not least of which during a pandemic when mass physical distancing has made it difficult to fully realize many important values (aside from physical privacy) A different orientation is needed at this moment As Flanagan Howe and Nissenbaum (2008) conceptualized in 2008 we should take a ldquovalues in designrdquo approach to DCTTmdashan approach that designs a broader range of values such as those enumerated above into technology
This approach requires a wider ethical lens through which to ex-amine DCTT and requires hard but important work to appropriately balance competing interests within technology architecture For example there is value in technology providing users the option to collect their location history and share it with public health professionals in order to advance the public health response increase system efficiencies (eg by contributing information that can lead to better data processing) and reduce the burden on essential workers For some this might be an ex-pression of autonomy solidarity or patriotism At the same time there is value in further advancing autonomy by designing technology to allow individuals to control what data about them are collected and shared
Justifying a Middle-Ground Approach to DCTT
We ought to embrace a DCTT that has a default of interoperability and privacy protection but that does not stop there Triggering events such as entry of a positive test result or receipt of a notification that one was proximate to someone who tested positive could for example generate a push notification that users can acknowledge in order to permit transmis-sion of potentially useful location data to public health authorities This could be accompanied by an explanation of the value of the information and relevant restrictions on its use
At this point it is worth reiterating that manual contact tracingmdash
52 Digital Contact Tracing for Pandemic Response
which involves collecting information from people whorsquove tested posi-tive and their contactsmdashincludes collection of personal information and potentially embarrassing or sensitive data about the places theyrsquove been and the people theyrsquove had contact with Manual contact tracing efforts use these data to uncover ongoing transmission provide useful informa-tion tailored to the individual and enable isolation and quarantine as necessary
It stands to reason that if these forms of data can be collected by a DCTT and provided to public health authorities in a maximally secure and voluntary way (with clear rules regarding authorized uses) this may amplify public health authoritiesrsquo manual contact tracing efforts For ex-ample location data from DCTT could help jog peoplersquos memories about where theyrsquove been and fill in memory gaps This is especially relevant given the long period of infectivity of SARS-CoV-2 which begins before people are symptomatic and therefore before they are aware they are infected (Ferretti et al 2020) Location data might reveal that a COV+ person was at a restaurant at an exact time and date which could be followed up by contact tracers who could alert the public or use other measures to reach those who were also present in the restaurant at the same time In other disease contexts (see Furlanello et al 2002 Dredze et al 2013 Eckhoff and Tatem 2015 Fraccaro et al 2019) geolocation data have demonstrated some potential to support epidemiology and dis-ease surveillance with technical cautions regarding accuracy and the like (Beukenhorst et al 2017)
These benefits are currently speculative for DCTT At present pro-viding public health authorities with large amounts of data on cases and potential case contacts will be useful only if there is sufficient public health capacity to follow up on these data In addition there is a risk of low-quality data from DCTT flooding the system Investigating poten-tial case contacts identified by a DCTT may distract them from other important efforts and at some point overwhelm public health capacity altogether Whether and to what extent data from DCTT will benefit contact tracing efforts is unknown pointing again to the importance of continuously collecting high-quality evidence about DCTT
Nevertheless what would enable the most flexible and potentially robust public health response is to design DCTT so that restricted data sharing is possible From an ethics perspective the collection and use
Ethics of Designing and Using DCTT 53
of sensitive data in manual contact tracing efforts (described above) is typically seen as ethically justifiable so long as there is sufficient public health benefit and need Thus wouldnrsquot it seem appropriate from both a public health and ethics perspective to design DCTT systems to enable similar data to be shared with public health authorities when and if there is ethical justification for sharing them
Why instead do so many advocate that DCTT should be designed as a ldquominimalrdquo system when this arguably ties the hands of public health and individual users and precludes the collection of data that public health authorities (and indeed many other apps on our phones) typically collect We here consider and appraise some of the reasons that may motivate individuals and groups to argue for minimalistic positions
1 Proponents of minimal systems may believe that such systems will be most
widely adopted Some groups have maintained that only these systems will earn and maintain public trust and be widely adopted (Simpson and Conner 2020) For example the previously referenced open letter (ldquoJoint Statement on Contact Tracingrdquo 2020) asserts ldquoSome of the Bluetooth-based proposals respect the individualrsquos right to privacy whilst others would enable (via mission creep) a form of government or private sector surveillance that would catastrophically hamper trust in and acceptance of such an application by society at large It is crucial that citizens trust the applications in order to produce sufficient uptake to make a difference in tackling the crisis It is vital that in coming out of the current crisis we do not create a tool that enables large scale data collection on the population either now or at a later time Thus solutions which allow reconstructing invasive information about the population should be rejected without further discussionrdquo
Response While it is true that public trust in and acceptance of DCTT is essential for its success there is insufficient evidence that public trust would be threatened by a DCTT system that has the capacity to collect location data and enable voluntary sharing of those data with public health authorities A contrasting perspec-tive is that maintaining public trust requires maintaining public confidence that the DCTT system is providing useful information is benefiting and not harming individuals and is advancing the
54 Digital Contact Tracing for Pandemic Response
public health response (Leprince-Ringuet 2020) From this per-spective a system that is less well integrated into the broader public health response or that generates a higher rate of false positives (as some suggest decentralized approaches might (Fraser et al 2020)) may fare worse when it comes to maintaining public confidence and trust
2 Proponents may hold the view that minimal systems are harmless (or nearly
harmless) to individuals This is because individuals are anonymous none of their location data are gathered and none of their identifiable data are shared with anyone In contrast DCTT systems that collect and share identifiable data including location data may be seen as posing risks of harm to individuals
Response While minimal systems may be harmless (or nearly harm-less) from the perspective of protecting privacy they may not be harmless from the perspective of public health if they generate system inefficiencies through producing too many false positive or false negative contacts Aside from presenting a challenge for public health professionals false positives could also harm individ-uals If users receive a large volume of automated messages alert-ing them to proximity events will this cause distress Will a large volume of alerts cause users to become disengaged and stop using the DCTT or lose confidence in contact tracing more generally as a legitimate method of disease control Admittedly these are just potential harms and risks it is unknown the degree to which they will materialize The point is that privacy-related harms are not the only relevant harms to individuals that we should consider when assessing DCTT
We acknowledge the risk under a middle-ground DCTT of data being used in ethically unjustifiable and harmful ways For exam-ple it would be against the principles and recommendations artic-ulated in this report for data to be sold or monetized by technology companies or others for corporate gain and this misuse of data would be more intrusive if the data were potentially identifiable What makes it ethically justifiable to take this risk is the compen-sating benefit of allowing the most flexible and robust public health
Ethics of Designing and Using DCTT 55
response during the pandemic but this alone is not sufficient The risk of inappropriate uses must be reduced by ensuring stringent requirements for data security and access as well as clear legal protections and recourse for any violations (as discussed further below)
3 Proponents may believe that DCTT systems should not collect location data
as this would be too intrusive and of insufficient value Some proponents of PPPT systems maintain that recording proximity events is sufficient and data relating to usersrsquo movement and location should not be collected (Ingram 2020) The thought may be all we need to know is whether two individuals came into close enough contact for viral transmission to have occurred we donrsquot need to know where or when this contact occurred and there is no need to collect and store usersrsquo location data
Response This conclusion might be too hasty As discussed above there is potential (though unproven) benefit to providing public health authorities with location data Location data could help jog peoplersquos memories about where theyrsquove been provide more context for understanding the nature of ldquoproximity eventsrdquo captured by the DCTT and allow public health authorities to quickly define a category of individuals who may be at risk Collecting location data from cases is what public health authorities do on a regular basis following best practices for manual contact tracing
In addition many peoplersquos location data are currently gathered by apps on their phones and used for various purposes such as to provide more accurate navigation to offer entertainment or to improve services Many are willing to accept these capabilities because they provide some value in return Why not allow DCTT to also collect these data so that the data are available for users to share with public health officials who can then do their work more effectively and refine their understanding of how the disease transmits If many are willing to have these data used to find a bet-ter route home why not let individuals share these data to support the effort to save lives
56 Digital Contact Tracing for Pandemic Response
4 Proponents may hold the view that minimal systems pose little or no threat
to individual autonomy whereas systems that collect identifiable data and
integrate public health do pose a threat to individual autonomy For exam-ple they may worry that use of DCTT could be mandated and not a voluntary choice and in this circumstance mandatory use of minimal DCTT would be less intrusive risky and privacy violating Another worry might be that itrsquos theoretically possible that DCTT could share individualsrsquo data with public health authorities without usersrsquo full understanding if the technology does not even gather identifiable data then itrsquos not possible for these data to be shared without the individualrsquos consent
Response We discuss the importance of appropriately designed disclosures and consent below as well as the high bar that would need to be met to ethically justify mandatory use At this time mandated use of DCTT by states or institutions is not justifiable given uncertainty about potential harms and benefits Users should have a meaningful opportunity to review and understand infor-mation about the specific technology and its uses and to consent Assuming that individuals are not required to use DCTT and that they provide consent to using it designing DCTT to make data collection and sharing possible is the design choice that maximizes individual autonomy because it provides individuals with options they may value
Individuals may wish to share their data with public health au-thorities for both self-interested and altruistic reasons For exam-ple someone who has tested positive for SARS-CoV-2 and enters this test result into an app may wish to be connected to public health authorities in order to be provided with needed information resources and support She may wish for public health authorities to be provided with her phone number in case they need to reach her to provide additional information Further someone who has been alerted by an app that he had a ldquoproximity eventrdquo with a person who has tested positive for SARS-CoV-2 may wish he had location data to share with public health authorities in order to help ascertain whether this event is a cause for concern or whether it is likely a false positive (eg he and the COV+ person were sepa-
Ethics of Designing and Using DCTT 57
rated by a wall) Someone who tests positive for the virus may also wish to share their location history with public health authorities in order to be as helpful as possible to the overall public health re-sponse by facilitating de-identified aggregate analyses that identify locations of higher transmission or contribute to refining overall understanding of the disease and pandemic
5 Concerns about ldquosurveillance creeprdquo and the long-term downstream effects of
digital contact tracing system may also motivate embrace of minimal DCTT Digital contact tracing technology that collects identifiers and loca-tion data and has the capacity to share them with public health au-thorities may represent a massive and concerning increase in govern-ment surveillance of the public It might be feared that the use of this surveillance capacity in the COVID-19 response sets an unwelcome precedent for future use in other contexts Designing DCTT as min-imal systems may be a way to minimize the risk of surveillance creep and to minimize the harms associated with potential future uses of the technology
Response Surveillance creep is a serious concern To guard against surveillance creep protections should be put in place to ensure that only those data that are necessary and relevant for the public health purposes at hand are collected and used and data should be kept only for the period of time needed for those public health purposes In the face of these concerns it is important to emphasize that widespread use of DCTT in the COVID-19 response is justi-fied by the exceptional circumstances of the current pandemic and their use in this context does not imply that future public health use is ethically appropriate without significant public debate (eg use in seasonal flu surveillance efforts) Future use will require in-dependent justification Use of DCTT in other contexts (eg law enforcement or immigration enforcement) is also presumptively unethical
All in all the arguments that DCTT should be designed as a minimal system are not convincing Rather DCTT should be developed through a ldquovalues in designrdquo approach with a core set of features that protect pri-
58 Digital Contact Tracing for Pandemic Response
vacy with enough flexibility to be used differently depending upon local conditions evolving evidence and individual preferences What kind of digital contact tracing system will strike the right balance between public health goals and other considerations will depend upon circumstances For example whether it is even beneficial to provide public health au-thorities with volumes of data about potential contacts of COV+ people will depend in part upon whether they have the capacity to make good use of those data This will vary from location to location and will change over time
Recommendations
bull Technology companies should not alone control the terms condi-tions or capabilities of DCTT nor should they presume to know what may be acceptable to members of the public
bull A ldquovalues in designrdquo approach to development of DCTT should be adopted (Flanagan Howe and Nissenbaum 2008 Knobel and Bowker 2011) Robust public- and user-engagement activities should be pursued to identify and incorporate to the extent pos-sible a range of values into the design of the technology These values may include privacy but also autonomy efficiency equity or others Technology design should reflect an appropriate balance and prioritization of identified values
bull Technology design should not be static but rather it should be ca-pable of evolving depending upon local conditions new evidence and changing preferences and priorities
bull DCTT should be designed to have a base set of features that protect privacy with layers of additional capabilities that users may choose to activate An initial default should be that user location data are not shared but users should be provided with easy mechanisms and prompts to allow for opting-in to this capability with encourage-ment to the public if and as it is shown to be critical to achieving public health goals
Ethics of Designing and Using DCTT 59
Policy Positions to Advance Widespread Use of Digital Contact Tracing Technologies
The public health value of a DCTT depends in part on the number of people who use it This section concerns broad public policy positions that relate to the widespread adoption of DCTT What are ethical means of encouraging or securing widespread adoption of DCTT systems Un-der what circumstances would it be ethical to mandate their use or incen-tivize their use What enforcement challenges exist
Mandating Use
Digital contact tracing has occurred without the publicrsquos explicit volun-tary agreement in some countries such as China and Israel In others use has been voluntary (Valentino-DeVries Singer and Krolik 2020) For example Singapore adopted an app that the public could use on a vol-untary basis and approximately 20 of the population has downloaded and used it Norway has recently launched a contact tracing app that was downloaded by roughly 30 of the population in the first week that it was made available In the United States many advocates and researchers have argued that use of digital contact tracing tools must be fully volun-tary this is the dominant perspective
There are numerous ways that DCTT could be put into use without user choice For example as has been done in Israel location data from mobile phones could be collected and used by the government without usersrsquo consent Use of an app could be formally mandated as a precon-dition for returning to work or school or even further to control entry into a facility or onto transportation such as airplanes through scanning of a QR code to demonstrate personal exposure levels (Gan and Culver 2020)
While these approaches are hard to imagine in the United States some contend that mandatory use of digital contact tracing tools could be ethical and may even be ethically required Mandating use of digital contact tracing tools could in theory vastly increase the effectiveness of digital contact tracing systems and thus may save more lives and allow states to lift lockdowns sooner or avoid reimposing lockdowns in the future Canca (2020) argues that use of privacy-by-design digital contact
60 Digital Contact Tracing for Pandemic Response
tracing tools should be mandatory because the use of these tools will be nearly harmless if there are sufficient privacy protections In addition mandatory use of DCTT that embraces these principles is significantly less intrusive at the individual level than manual contact tracing which involves the collection of personally identifying and potentially sensitive data In this light it could be argued that such mandates are actually pref-erable from the perspective of both public health and individual liberty insofar as they reduce the likelihood of ldquostay at homerdquo orders which are a severe limitation of individual liberty
Nevertheless mandated use of DCTT systems faces considerable ob-stacles For example people may not adhere to the mandate by simply leaving their phone at home thus preventing their activities from being tracked Even more harmful would be if people react to a mandate and a perceived violation of liberty and privacy by employing location and Bluetooth spoofing software to shield their real contacts behind a screen of misinformation The introduction of this misinformation into a contact tracing effort might severely undermine its effectiveness The possibility of nonadherence also raises the issue of enforcement would high rates of nonadherence be permitted or would enforcement be attempted (if even possible) Perhaps more important should the technology not deliver the hoped-for benefits having mandated the use of an unproven technology could result in a loss of public trust in the technology the entity instituting the mandate and potentially the larger public health response (Bernstein et al 2019)
Mandatory DCTT could also be used to enforce quarantine restric-tions and stay-at-home orders for those who are COV+ or are determined to be at heightened risk The use of DCTT in enforcement activities raises a number of ethical (and legal) issues that are beyond the scope of the present analysis In particular individuals have a heightened interest in personal privacy if their data can be used to restrict their freedom of movement and other civil liberties At a minimum stringent procedural protections would be required to ensure that the data collection is fair and unbiased and that DCTT users are provided with adequate informa-tion in advance about how their data may be used
Mandatory use policies for DCTT must therefore convincingly ad-dress a number of questions including
Ethics of Designing and Using DCTT 61
bull Is the technology designed to meet an important and unmet public health need
bull Is there sufficient evidence to suggest that the technology will be effective at serving its purpose
bull Is the outbreak characterized by sufficiently severe morbidity and mortality and a high rate of disease transmission
bull Are there other less autonomy-restricting or less risky alternatives to widespread mandatory use of DCTT
bull Is it possible and likely that a sufficient number of individuals will comply with a mandate
bull Can inequities in the burdens and benefits of the mandate be suffi-ciently addressed through social protections and countermeasures
bull Can enforcement and enforcement discretion be implemented in a manner that is consonant with fundamental rights
bull Will those subject to the mandate interact closely with a population that is at high risk of morbidity or mortality if they contract the virus
bull Is it possible to mandate use and remain consistent with important ethical and legal principles
These questions would need to be satisfactorily addressed and explicitly documented by any decision maker considering mandatory use includ-ing government officials institutional leaders and employers Particularly important is the need to identify reliable evidence that the DCTT would be effective and to ensure that the burdens and benefits of use are equita-ble and justifiable If use of a DCTT is a condition for returning to work or school in person those who refuse or are unable to use DCTT should not lose their jobs or positions as a result and adequate support should be in place for people who are asked to self-quarantine
Finally it is important to distinguish a mandate from a ldquopushedrdquo program installation or a default setting in an application which can be modified by users A mandate relates to a policy of required use whereas the pushed programs or default settings relate to the chosen architecture for download and operation of the application
62 Digital Contact Tracing for Pandemic Response
Incentivizing Use
Perhaps the most effective way to generate widespread adoption of DCTT in the United States is to offer incentives to individuals who choose to adopt and who properly utilize the preferred DCTT approach in a voluntary system External incentives may help ldquonudgerdquo populations toward desired adoption targets Given the importance of widespread use of DCTT modest incentives ought to be considered for DCTT in the US if and when there is sufficient evidence of the technologyrsquos utility Note that in other contexts studies have shown that the provision of some incentive leads to an increase in adoption or utilization of public health programs (Singer and Ye 2013 Lee et al 2014) Moreover even a relatively small incentive can achieve much greater rates of adoption with some studies demonstrating that the incremental adoption gain de-creases as the incentive gets larger (Thornton 2008 Gibson et al 2019) In the context of COVID-19 incentives that might be both effective and ethically acceptable could include a relatively small monetary token free or discounted mobile phone service for a period of time or credit to be used by means of a mobile phone
Not all incentives are ethically appropriate For example making access to lifesaving health care contingent on using a DCTT or making valuable disease information available only to DCTT users would not be ethically appropriate In addition incentives cannot be used to over-come otherwise ethically unjustifiable technology design for example they should not be used as an offset for providing personally identifiable health information to other users
Importantly incentivization schemes must be kept distinct from man-dates as the latter require greater ethical justification To offer an incen-tive is to offer something of actual value to individual participants over and above what they are reasonably entitled to at baseline For example making a return to work contingent on using DCTT is not offering an incentive but instead imposing a mandate and it would have to be justi-fied as a mandate
In the context of COVID-19 it is also necessary to recognize that there is an inherent ldquoincentiverdquo behind the technologymdashthat is the prom-ise of more lives saved faster pandemic recovery and the reduction or elimination of blanket physical distancing Effective public communica-
Ethics of Designing and Using DCTT 63
tion of these goals if and when there is sufficient confidence in the tech-nology is important
Encouraging Use
Another important approach to increasing use of DCTT in the United States is for trusted leaders to encourage their use Community leaders public figures health care professionals and other respected individuals who have the publicrsquos trust and goodwill could be enlisted to commu-nicate with the public about DCTT and encourage its use drawing on notions such as communal responsibility solidarity and so on These en-couragements could be combined with other approaches (eg small in-centives) to optimize reach while continuing to respect individual choice
Recommendations
bull DCTT use should not be mandated at this time given uncertainty about potential harms and benefits Additional technology user and real-world testing is needed
bull Incentives can be a useful complement to encouragements how-ever any incentives for users to install and use DCTT must be equitable should not be coercive and should align with effective use of the technology (ie they should not incentivize downloading an app but then leaving onersquos phone at home)
bull Trusted leaders should be enlisted to communicate effectively with the public about DCTT and encourage its use should the technol-ogy demonstrate some potential The limits of knowledge regard-ing effectiveness should also be explained along with what will be done to improve technological capabilities as understanding evolves
Disclosure and AuthorizationConsent
In deciding whether to use DCTT voluntarily individuals must be suf-ficiently informed both through broad coordinated public engagement campaigns and individual-level disclosures and there must be a meaning-ful mechanism for users to consent It is important to recognize that while
64 Digital Contact Tracing for Pandemic Response
informed consentmdashwhich is characterized by detailed consent forms and requires a witnessed signaturemdashis the standard for most research and clinical care encounters (Faden and Beauchamp 1986) it is not typically the standard for public health disease surveillance In the public health context other relevant protections (such ethics training for public health professionals and strict data handling and confidentiality requirements) are in place and there is a strong public health interest in collecting the relevant data A more limited role for consent has been recommended for public health surveillance based on a reciprocal obligation of members of society to contribute to a ldquocommon goodrdquo and particularly in the con-text of a pandemic practical considerations such as time constraints and exigencies such as increasing morbidity and mortality (WHO 2017)
Under current circumstances given that (1) many individuals have time and capacity to consent (2) DCTT is being considered as part of plans for longer-term restabilization (3) DCTT is not a familiar part of our public lexicon (4) remote consent disclosure and authorization can be easily embedded in DCTT systems (Moore et al 2017) and (5) there are justifiable public deficits in trust with respect to various government and corporate actors handling potentially personal digital information a strong ethical case can be made for requiring a carefully crafted version of what is sometimes referred to as simple consent Simple consent consists of basic disclosure and voluntary agreement or authorization (Ali et al 2017) Three questions then arise
1 What information should be disclosed to potential users of DCTT
bull Information disclosed might include
deg Entity responsible for the technology
deg Its purpose
deg How it works (in lay terms)
Some participatory disease surveillance systems (eg Flu Near You) have received for-
mal ldquowaiversrdquo of consent requirements from institutional review boards (IRBs) in the US
As they undergo development these digital surveillance systems often straddle a line be-
tween public health surveillance and research hence the frequent need or desire to obtain
ethical review by an IRB (Ali et al 2019)
Ethics of Designing and Using DCTT 65
deg What users need to do
deg Any user options eg
Sharing geolocation data with public health authorities when that would facilitate a defined public health goal
Sharing de-identified metadata with technology develop-ers (for system enhancement)
deg User rights
deg How data will be handled
What data are collected
What data are shared (and how and with whom)
Purposes for which data can be used and not used
How data are secured and protected
Whether and what data will be retained (or will be deletable)
deg Potential benefits and any known risks
deg How to obtain answers to questions about the technology and public health response
2 How should this information be presented
Information should be presented leveraging eConsent models that are more accessible than long ldquoclickwraprdquo disclosures typical of mobile apps (Iwaya et al 2019) For example a simple open-source smartphone con-sent module that has been developed by Sage Bionetworks for research uses could be adapted to the public health surveillance context and to DCTT (Doerr Suver and Wilbanks 2016)
bull Formatting recommendations include (cf Doerr et al 2016)
deg simple and straightforward information
deg deliberately organized content
deg multimodal learning (eg visual audio written)
deg accessibility for disabled users
66 Digital Contact Tracing for Pandemic Response
deg multilingual text
deg engagement through interaction (eg swiping to navigate forward and backward)
bull The same simple information should be made publicly available via multiple other platforms (eg on websites in newspapers over social media)
bull More detailed disclosures should be made readily accessible to those who wish to learn more with no hidden surprises
3 How should users signal that they agree to the details specified in disclosures
Opt-in Models
Opt-in models are those that through an affirmative act such as clicking a button users would indicate their intention to use a DCTT This ap-proach is consistent with other app downloads where app details and privacy policies are made available through a download page and users are required to affirmatively click a button to install an app Once in-stalled some apps further alert users to particular ways in which phone capabilities or data will be used with some permitting selective toggling (opting-in or opting-out) of certain features With DCTT apps in addi-tion to disclosures provided on a download page the user could be guided through a simple interactive module embedded in the app (such as is described above) in order to increase the chance of meaningful exposure to important information about the technology and how data will be handled At that point any user options such as those itemized above could be described and choices made
Opt-out Models
There are at least two different ways in which the term ldquoopt-outrdquo has been used in this context The conventional use of the term ldquoopt-outrdquo is characterized by an act which signals an individualrsquos intention to decline something that would have otherwise occurred without intervention A few others have used the term to refer to ldquorevocation of consentrdquo for example the United States COVID-19 Consumer Data Protection Act of 2020 Senate bill (S3663) would establish a default opt-in positionmdash
Ethics of Designing and Using DCTT 67
requiring ldquoaffirmative express consentrdquo for collection and use of prox-imity and other related datamdashand refers to individuals having a right to later revoke their consent through an ldquoopt-outrdquo The latter use of the term is not our focus here
Given this a DCTT app that is voluntarily downloaded through an affirmative act would be difficult to characterize as an opt-out approach This leaves more passive surveillance systems that rely on automatic in-stallation of self-activating technology onto users phones There are a range of views among the authors of this report about the value of an opt-out approach for DCTT with some arguing for an opt-out approach on grounds that it might increase coverage and would be ethically acceptable if accompanied by similar disclosures as above to ensure users are aware of the technology and data uses (Mello and Wang 2020) This approach would present users with a mechanism to opt-out if they wish which should be reasonably easy to effectuate Under these circumstances as noted above an ldquoopt-outrdquo would not be synonymous with mandating use of the technology
Others among the authors argue that there is reason to believe that opt-in approaches may be able to sufficiently achieve desirable levels of utilization relative to opt-out approaches Unfortunately data related to opt-in versus opt-out models of DCTT are very limited One recent sur-vey (Altmann et al 2020) found that across five countries (UK Germany France Italy US) slightly more people reportedly would download an app under an opt-in system (748) than would keep an app on their phone under an opt-out system (677) Moreover when US respondents were directly asked which approach they would prefer 60 indicated a preference for opt-in This remained true across various demographic variablesmdashgender region political affiliation lockdown status and other characteristics Whether actual behaviors would align with anticipated behaviors in the context of DCTT remains an unanswered question that should be carefully studied under real-world conditions There are a range of important empirical questions regarding how much and what kind of impact (positive or negative) various types of defaults might generate for public health and for different mobile phone user groups including vul-nerable and marginalized users
Opt-out models for app authorization may encounter greater legal
68 Digital Contact Tracing for Pandemic Response
and political challenges especially if the COVID-19 Consumer Data Pro-tection Act of 2020 (S3663) the competing Public Health Emergency Pri-vacy Act (S3749) or another similar bill is enacted in the United States Both of these standing bills require affirmative opt-in consent Opt-out approaches also risk negative reactions from some mobile phone users a small number of whom may go so far as to intentionally interfere with data because of the perceived intrusiveness of an automatically installed tracking platform (Dixit 2020)
Given these considerations and the apparent willingness of a large portion of the population to opt-in to use DCTT an opt-in approach to authorization should be instituted to accompany initial DCTT rollout The feasibility and value of opt-out approaches should continue to be evaluated informed by what is technologically possible what local as-sessments of benefits and harms of the technology reveal over time and our evolving understanding of the degree to which an opt-out approach is likely to increase or decrease utilization Opt-out approaches should not be precluded
Recommendations
bull A clear and concise module consisting of basic disclosure and vol-untary authorization should be developed to accompany DCTT This module should not take the form of ldquoclickwraprdquo terms of service or end-user agreements but rather provide only essential information necessary for an individual to make a decision More detailed disclosures (such as FAQs in plain language) should be made easily accessible to those who wish to learn more with no hidden surprises
bull An opt-in approach to authorization should be instituted to ac-company initial DCTT rollout The feasibility and value of opt-out approaches should continue to be evaluated informed by what is technologically possible what local assessments of benefits and harms of the technology reveal over time and our evolving under-standing of the degree to which an opt-out approach is likely to increase or decrease utilization among different populations Opt-out approaches should not be precluded
Ethics of Designing and Using DCTT 69
Promoting Equity and Fairness
Digital contact tracing technologies should be designed and used in ways that as far as possible promote an equitable distribution of benefits and burdens DCTT should be deployed in a manner that does not propa-gate preexisting patterns of unfair disadvantage or distribute harms and risks unfairly throughout the population For example communities with lower rates of technology and data access may benefit less from DCTT Special attention must be paid to communities that experience preexisting health disparities and to those that are being hardest hit by the pandemic
Digital Disparities
In the United States February 2019 data indicate that approximately 80 of the population are smartphone users (Pew Research Center 2020) though rates of mobile phone use are significantly lower among people over age 65 (53) people with any disability (58 2016 data) (Anderson and Perrin 2017) people with less than a high school edu-cation (66) people who earn less than $30000 per year (71) and people who live in rural areas (71) As a result these populations and communities may use DCTT in lower numbers thereby lessening the effectiveness of DCTT and the likelihood of benefit for these populations from such systems Moreover it has been reported that many older and less costly smartphones (roughly estimated at 10ndash20 of smartphones in the US) lack important capabilities required for the leading AppleGoogle platform to work (Bradshaw 2020) This is of special concern because some of the above groups that are less likely to own smartphones in general are also less likely to own newer smartphones with the needed capabilities Some within the above groups (eg people who are older and people identified as Hispanic African American or American Indian) are also disproportionately experiencing morbidity and mortality from COVID-19 (CDC 2020h)
One may argue that by using DCTT human and financial resources that would otherwise be spent on manual contact tracing will be pre-served and these resources can then be redirected to better meet the needs of those who are not otherwise being effectively served by the technology because of disparities or for other reasons This argument has intuitive
70 Digital Contact Tracing for Pandemic Response
appeal and should be taken seriously however it is unsettled whether DCTT will contribute sufficient efficiencies to the overall public health response to make it possible financially and logistically for manual ser-vices to be allocated in greater proportion to those who are unable to benefit from DCTT It is entirely possible that at least in the short-term DCTT may introduce new inefficiencies due to unintended consequences or the need for public health officials to follow up many more contacts One possible mitigation to the challenge of digital disparitymdashthough it does not solve the underlying challenge of ensuring net efficiency across systemsmdashmight be to provide mobile phones or other devices and data packages to those who would otherwise be left out
Disparate Risk of Harm from Surveillance and Data Gathering
Ensuring wide digital coverage does not however resolve other equity concerns It is important to consider that some populations may experi-ence greater harm and fear of harm from having their data collected For example some groups such as African Americans Hispanic Americans Muslim Americans and undocumented immigrants have more reasonable fear of their data being handed over to law or immigration enforcement and some groups have lower levels of trust in public health due to past injustices (CSM 2017 Pew Research Center 2017 Rodrigues et al 2018 Auxier et al 2019) Any data gathered by DCTT should be used solely for public health purposes Efforts should be made to assure members of these and other communities that their data will not be misused or made available to those outside of a public health context In addition if DCTT are used in the current pandemic this should be with the understanding that future use of DCTT in other contexts (eg law enforcement or im-migration enforcement) is presumptively unethical
Some preliminary polling related specifically to DCTT emphasizes the complexity of the challenges faced and the need for deeper public engagement (Anderson and Auxier 2020) The polling results suggest that people who identify as African American or Hispanic are more likely than people who identify as White to consider government tracking of mobile phones as acceptable These findings like many others are difficult to in-terpret given background political polarization on the issue More direct engagement is required to better understand how different communities comprehend and experience DCTT and other forms of surveillance
Ethics of Designing and Using DCTT 71
Discrimination and Stigma
Stigma may result from an individual being identified as COV+ or a neighborhood or establishment becoming identified as a ldquohotspotrdquo as a result of numerous COV+ people living in that area or having visited that establishment In particular certain groups may suffer more as a result of being associated with COVID-19 such as the well-documented blame that has been directed toward Chinese people (and broadly East Asian communities) or the communities that are disproportionately likely to contract the illness (Devakumar et al 2020) When identifiable lo-cation data are made public as has been the case in South Korea per-sonal and private information were revealed Furthermore businesses in South Korea that were identified as having patrons who tested positive for COVID-19 have suffered economic losses and stigma (N Kim 2020)
To avoid the stigma and potential discrimination that can result from being identified as COV+ DCTT must never make data publicly avail-able that could be used to identify persons who have tested positive Safe-guards must be in place to ensure that any identifiable data that may be gathered for public health purposes are protected If DCTT data are used to provide heat maps to the public of locations that COV+ individuals frequently visit so as to provide representations of geographic risk or for other reasons it is essential that care be taken to avoid unfairly distrib-uting further economic burdens or other stigmatizing and discriminatory outcomes
Recommendations
bull A commitment to equity means a commitment to ensuring that the benefits and burdens of DCTT are distributed fairly Public engagement is an important tool for assessing impact and to rectify inequities
bull States localities and institutions that recommend widespread use of DCTT should provide technology (eg mobile phones Blue-tooth devices) and free data packages to those who desire but lack access to these devices
bull If there are lower rates of adoption of DCTT systems in some identifiable communities public health authorities should iden-
72 Digital Contact Tracing for Pandemic Response
tify ways to compensate For example directing more non-DCTT resources and efforts toward those communities to meet specific needs that are elsewhere being supported by technology
bull If maps are generated based on DCTT to provide the public with the locations that COV+ individuals have visited steps must be taken to minimize the stigma and potential financial losses that could result from being identified as a hotspot
Instituting Transparent Governance and Oversight
DCTT must be developed with an eye toward both present and future implications Transparent and publicly trustworthy management gover-nance and oversight of DCTT technology and data is both a near- and long-term necessity We face significant uncertainties DCTT technologies are rapidly developing Their risks capabilities effectiveness and down-stream implications are not yet well understood
Concerns about ldquoSurveillance Creeprdquo
Significant concerns have been expressed by privacy advocates (Guari-glia 2020) and in the popular press (Giglio 2020) about what is known as ldquosurveillance creeprdquo Their worry is that state and corporate actors will use new surveillance technologies capacities and permissions well beyond the purposes for which they were initially justified to the public and beyond the time when they are useful for the COVID-19 pandemic
Surveillance creep should be guarded against Only those data that are necessary and relevant for the public health purposes at hand should be collected and used and data should be kept only for the period of time needed for those public health purposes Data should be used only for public health purposes
Any use of DCTT during the current pandemic would be justified by the circumstances of this pandemic and its use in this context does not set a precedent for future public health use (eg use in seasonal flu surveillance efforts) Future use will require independent justification Use of DCTT in the future in other contexts (eg law enforcement or immi-gration enforcement) is presumptively unethical
Broadly speaking efforts should be made to generate public aware-
Ethics of Designing and Using DCTT 73
ness and consensus that use of DCTT in COVID-19 efforts does not imply that future use is justifiable However generating this public aware-ness may be particularly challenging given the complexity of the informa-tional environment where public debate ranges from legitimate concerns about surveillance creep to conspiracy theories regarding the origins of the COVID-19 pandemic (Muller 2020) This means authorities bear spe-cial obligations to be clear on how they plan to use the technologies what oversight mechanisms will be employed to address potential abuse and how they intend to publicize the conditions under which programs will be terminated making sure they are followed
Oversight and Ethical Review
We are rapidly gaining knowledge about SARS-CoV-2 and COVID-19 but we still have essential gaps in our understanding In the United States public health responses including DCTT will generally be developed and coordinated by individual states regional consortia and associations (Reston Sgueglia and Mossburg 2020) Good governance in this con-text requires transparency and the creation of oversight bodies with the appropriate expertise and representation to allow nimble and effective responses while serving as trusted representatives
To address the range of ethics-related concerns about the design and use of DCTT digital surveillance oversight committees should be estab-lished perhaps at a state level and with a platform for national coordina-tion These committees can provide ethical and regulatory review prior to and concurrent with widespread use of DCTT These committees should be composed of a diverse group of experts capable of evaluating a DCTT system locally including members of communities that experience higher rates of digital disparity
When assessing the design and use of digital contact tracing systems these committees (and the public more widely) should consider not only the risks and benefits accrued during the COVID-19 pandemic but also implications for the future What kind of precedent might use of these technologies during the current pandemic set for future use capabilities in other infectious disease outbreaks or in other social contexts (eg law enforcement) How can we navigate safe use of these technologies in a way that preserves public trust in them and enables the possibility of future beneficial use
74 Digital Contact Tracing for Pandemic Response
As a start it should be emphasized that principles offered in this and other guidance documents do not apply only during the pandemic Future efforts to advance DCTT capabilities during quieter times should make every effort to follow them
Recommendations
bull Digital surveillance oversight committees should be established expeditiously with diverse and qualified membership to provide ethical and regulatory review prior to and concurrent with wide-spread use of a DCTT system
bull Understandable and publicly accessible rules must guide the col-lection access control use storage and combination of data by government authorities public and private institutions and other parties such as public health researchers
bull Only those data that are necessary and relevant for the public health response to COVID-19 should be collected and used
bull Identifiable data should be kept only for the period of time needed for the public health response to COVID-19
bull Identifiable data collected as part of this response should not be shared with anyone other than the relevant public health authorities without additional specific informed consent of individual users
bull Before a government or institution adopts a digital contact tracing program they should state the conditions under which the digital contact tracing program will be terminated
bull Future use of DCTT to advance public health or other efforts (eg use in seasonal flu surveillance) would require independent justification DCTT designed for public health use should not be used by law or immigration enforcement
bull The principles offered in this guidance document apply both during and following the COVID-19 pandemic
75
The implementation of digital contact tracing technology (DCTT) is likely to implicate a number of US laws at both the federal and state levels This section focuses primarily on federal laws as these laws apply nationwide and generally preempt conflicting state laws A comprehen-sive assessment of the legality of any particular DCTT program would require case-specific analysis and attention to relevant state laws includ-ing any that specifically address DCTT which may soon exist in one or more states The analysis here is limited to the United States foreign and international laws will not be addressed
Many of the laws discussed in this section are privacy laws designed to protect individuals from the harms that may result from the unautho-rized or improper use of their personally identifiable information (PII) Under these laws legal concerns will generally be minimized if privacy protections are built directly into the DCTT technology (eg ldquoprivacy by designrdquo) As a general principle DCTT should be designed to collect and store only as much PII as is necessary to achieve the public health purpose Collecting only proximity data for example is likely to raise fewer legal concerns than collecting both proximity data and geolocation data Likewise creating aggregated anonymized or de-identified data will raise fewer legal concerns than using and disclosing PII
As we have argued elsewhere in this guidance document however the public health and societal crisis caused by COVID-19 may justify
Legal Considerations
FOUR
76 Digital Contact Tracing for Pandemic Response
greater encroachments on individual privacy than would otherwise be permissible Regardless of the type of data collected privacy concerns will be reduced if users are afforded the right to choose whether their PII is collected and how it is used and disclosed As such DCTT should gener-ally secure meaningful user consent before collecting PII a process which typically requires both disclosure of relevant information and agreement on the part of the user
Privacy concerns will also be reduced if the use of PII is strictly lim-ited to tracking and limiting the spread of SARS-CoV-2 The use of DCTT data for other purposesmdashsuch as commercial or law enforcement pur-posesmdashwould raise additional legal and ethical concerns In addition DCTT developers may be required to implement governance policies that ensure the secure storage of PII limit data retention periods require transparency about data sharing and maintain records of responses to data requests from government authorities
In short the legality of a DCTT program under current United States law will depend on a number of factors including what type of data is collected how the data are used and who may access them how user consent is obtained whether the entity collecting and using the data is the government or a private corporation the context in which data are collected (eg employment education or commercial) and which states have jurisdiction over the program
Privacy law in the United States unlike in other jurisdictions such as the European Union (EU) and Australia is generally sector-specific and limited in scope The result is a patchwork of protections that differ significantly depending on the entity that collects the data and the type of data collected For example under current law telecommunication carriers are governed by different privacy rules than mobile broadband providers Given the complexity of existing federal privacy law we be-lieve that it would be beneficial for the US Congress to enact new privacy legislation that is specifically tailored to the use of DCTT in response to COVID-19 Congress appears poised at least to debate such legislation a pair of bills recently introduced in the Senate and one in the House of Representatives would significantly restrict the collection of PII by digital devices for COVID-tracing purposes S3663 S3749 HR 6866 116th Cong (2020)
Legal Considerations 77
Data Privacy and Data Security Laws
Telecommunications
A DCTT provider that collects data from a userrsquos mobile phone may be subject to the privacy rules governing telecommunication carriers which are enforced by the Federal Communications Commission (FCC) The data protected under these rules are limited however to certain types of PII termed ldquocustomer proprietary network informationrdquo (CPNI) More-over the rules generally apply only to telecommunications carriers and interconnected VoIP (Voice over Internet Protocol) providers
In particular under section 222 of the Communications Act of 1934 47 USC sect 222 and the implementing regulations of the Federal Com-munications Commission (FCC) telecommunications carriers and VoIP providers must establish and maintain systems designed to ensure that they adequately protect their subscribersrsquo CPNI and they are generally restricted from using or disclosing CPNI without the customerrsquos consent (unless the use of disclosure is needed to provide the services subscribed to by the customer) If customer consent is sought to use or disclose CPNI individual notice must be provided to the customer and such notice must provide sufficient information to enable the customer to make an in-formed decision as to whether to permit the requested use or disclosure
CPNI is individually identifiable information that carriers and pro-viders have collected about their customers including phone numbers called and the frequency duration and timing of such calls Of most relevance to DCTT a recent FCC Notice of Apparent Liability asserted that user geolocation data collected by mobile phone network carriers qualify as CPNI under sect 222 and related rules 35 FCC Rcd 1785 (2) (2020) Pursuant to this notice the FCC fined T-Mobile for selling to third parties location data that were derived from the communication between the mobile phones of T-Mobilersquos customers and nearby network signal towers (The FCC also levied fines against ATampT Verizon and Sprint on the same grounds (Valentino-DeVries 2020)) While the FCC has made its position clear that geolocation data are CPNI courts have yet to weigh in on the matter
Even if geolocation data are CPNI however the FCC can enforce sect 222 of the Communications Act only against telecom carriers and VoIP
78 Digital Contact Tracing for Pandemic Response
providers not against cable broadband and mobile broadband internet providers 47 USC sect 53(44) 47 CFR sect 93 In 2018 the FCC promul-gated a regulation stating that contrary to its prior position its sect 222 authority does not extend to cable broadband and mobile broadband internet providers Restoring Internet Freedom 83 Fed Reg 7852 (Feb 2 2020) (to be codified at 47 CFR pts 1 8 and 20) This regulatory shift was subsequently upheld by the DC Circuit Mozilla Corporation v Federal Communications Commission 940 F3d 1 (2019)
In addition to sect 222 the FCC has authority to regulate ldquocommon carriersrdquomdashincluding both telecommunication carriers and broadband internet providersmdashunder sect 201(b) of the Communications Act In the past the FCC has interpreted sect 201(b) to protect against ldquounjust and unreasonablerdquo privacy and data security practices with respect to custom-ersrsquo personal information beyond CPNI In 2016 the FCC promulgated a regulation asserting its authority under this interpretation However Congress overturned this regulation pursuant to the Congressional Re-view Act in 2017 SJ Res 34 115th Cong (2017) At present the extent of the FCCrsquos authority under sect 201(b) remains unsettled (Mulligan and Linebaugh 2019)
Consumer Protection
The collection storage release and transmission of digital user data in-cluding proximity contacts is more generally governed by the Federal Trade Commission (FTC) The FTC is an independent US law enforce-ment agency tasked with protecting consumers and promoting competi-tion across broad sectors of the economy (FTC 2020) The FTCrsquos primary legal authority with respect to consumer protection comes from Section 5 of the FTC Act which prohibits ldquounfair or deceptive acts or practices in or affecting commercerdquo 15 USC sect 45(a)(1) Note that the FTC and FCC have some overlapping authority to protect consumer privacy in the context of telecommunications (FCC and FTC 2017)
The FTC has interpreted Section 5 to require companies to be trans-parent and accurate about their collection of PII from consumers A com-pany may be found to have engaged in a deceptive practice if it fails to disclose that it is collecting user data or fails to disclose that it is sharing these data with third parties and to provide a general description of these third parties The FTC has used its authority under Section 5 numerous
Legal Considerations 79
times to discipline companies that purport in published privacy policies or other notices to provide protection for the privacy andor security of personal information yet fail to do so in practice For example the FTC may find it both ldquounfairrdquo and ldquodeceptiverdquo for a mobile app privacy policy to state that the app never discloses location information to third parties when in fact the app shares that information with the app developerrsquos service provider which in turn uses it to provide analytical data to the app developer that are used to create targeted advertising
The FTC does not use its Section 5 authority other than to protect consumers and generally does not consider ldquode-identifiedrdquo user data which are data that are not ldquoreasonably linkablerdquo to a consumer to be a subject for consumer protection In general data collected are not ldquorea-sonably linkablerdquo so long as the company collecting it ldquo(1) takes rea-sonable measures to ensure that the data are de-identified (2) publicly commits not to try to reidentify the data and (3) contractually prohibits downstream recipients from trying to reidentify the datardquo (FTC 2012)
Many states have laws that are similar to Section 5 prohibiting un-fair and deceptive acts and practices Both Section 5 and these similar state laws can be violated not only by misrepresentations (affirmative deception) but also by material omissions Thus a failure to inform an app user of the apprsquos collection of tracking data and the planned use and disclosure of those data could constitute a violation of these laws Com-panies providing DCTT apps should make sure that all such information is disclosed in the appsrsquo terms of use to which users must affirmatively agree
Childrenrsquos Online Privacy
Children who use DCTT may be protected by additional privacy protec-tions In particular collection of digital PII from children under the age of 13 is strictly regulated under the Childrenrsquos Online Privacy Protection Act (COPPA) (15 USC sectsect 6501ndash6505) Under COPPA PII includes ldquofirst and last name[] a persistent identifier that can be used to recognize a user over time and across different online services[] and geolocation infor-mation sufficient to identify street name and name of a city or town[]rdquo COPPA prohibits a website or online service from collecting personal information (including location information) from children under age 13 without obtaining verifiable parental consent Note that there may be an
80 Digital Contact Tracing for Pandemic Response
exception to this requirement for an ldquoinvestigation on a matter related to public safetyrdquo 16 CFR sect 3125(c)(6)(iv)
Electronic Surveillance
In addition to misuse of user data by DCTT providers another privacy concern is that a third party may be able to access sensitive PII that is collected and stored by a DCTT system without the userrsquos knowledge and consent There are a number of federal criminal laws however that would likely prohibit such unauthorized access to PII
In particular the Electronic Communications Privacy Act of 1986 (ECPA)mdashwhich includes the Wiretap Act (18 USC sectsect 2510ndash2522) the Stored Communications Act (18 USC sectsect 2701ndash2711) and the Pen Register Act (18 USC sectsect 3121ndash3127)mdashmakes it a crime to access elec-tronic communications without authorization Individuals who violate the ECPA face up to five years in prison and fines up to $250000 Victims are also entitled to bring civil suits and recover actual damages in addi-tion to punitive damages and attorneyrsquos fees for violations
Generally the access restrictions in the ECPA apply unless consent is given or if access is authorized by statute for law enforcement purposes For example an employer is generally forbidden from accessing an em-ployeersquos private emails However if consent is given in the form of an employment contract that explicitly authorizes the employer to access emails it may be lawful under the ECPA for the employer to access such information Along the same lines the ECPA would likely prohibit an employer from accessing contact tracing data on an employeersquos phone without the employeersquos consent However the ECPA would likely not prohibit duly authorized government public health officials from access-ing contact tracing data without consent
As its name suggests the Stored Communications Act (SCA) regu-lates access to communications at rest that is not in transit The SCA makes it unlawful to intentionally access a facility in which electronic communication services are provided and to obtain alter or prevent au-thorized access to a wire or electronic communication while it is in elec-tronic storage in such a system As such the SCA would likely apply only to centralized collection of contact tracing data
The Pen Register Act covers any ldquosignaling informationrdquo exchanged in a communication such as phone numbers The statute does not reach
Legal Considerations 81
the content of such communications however An expansive interpreta-tion of the Pen Register Act would cover Bluetooth ldquohandshakesrdquo as they are merely signaling information between devices which do not carry content See United States v Forrester 512 F3d 500 (9th Cir 2007) (find-ing that email headers and IP addresses are akin to pen registers and have no Fourth Amendment protection) Unlike the SCA there is no statutory exclusionary rule that applies when the government illegally uses a pen register trap and trace device Additionally there is no private cause of action against the government for violations of the Pen Register Act
State Data Privacy Laws
States have a variety of privacy laws and are increasingly seeking to reg-ulate the online collection of personal information and the use and dis-closure of such information To date most of these laws focus more on transparency and protection from unauthorized access than on restricted collection and use (except with respect to biometric information) seek-ing to ensure that individuals who use websites or online services such as mobile applications do so on an informed basis with respect to the privacy provided by those sites and services Two examples of such state laws are the California Online Privacy Protection Act (CalOPPA) and the California Consumer Privacy Act (CCPA) Both laws require notice to in-dividuals who use websites or online services such as mobile applications in order to ensure that users are informed about the privacy of personal information collected by those sites and services (The CCPA also applies to data collection off-line) Both laws treat IP addresses and location data as types of potentially identifiable personal data and so would very likely apply to DCTT apps used by California residents
CalOPPA requires that the operator of any website mobile appli-cation or other online service (ldquoSiterdquo) post a privacy policy on the Site disclosing certain information regarding the Sitersquos collection use and dis-closure of PII CalOPPA applies to any Site that is accessible to California residents The required disclosures are not onerous and would apply only to collection of data that are identifiable to an individual person (but depending on who collects the data location data together with a device identifier are identifiable to the user)
The CCPA requires that any entity qualifying as a ldquobusinessrdquo provide its ldquoconsumersrdquomdashdefined as lawful residents of Californiamdashwith specific
82 Digital Contact Tracing for Pandemic Response
disclosures about the businessrsquos collection use and disclosure of personal information Importantly the CCPA applies only to for-profit businesses that meet certain thresholds of revenue or access to consumer informa-tion A public health agency or a nonprofit organization would not be subject to the CCPA Cal Civ Code sect 1798140(c)
The CCPA defines ldquopersonal informationrdquo as ldquoinformation that iden-tifies relates to describes is reasonably capable of being associated with or could reasonably be linked directly or indirectly with a particular consumer or householdrdquo The statute provides a nonexclusive list of po-tential identifiable personal information including ldquogeolocation datardquo In accordance with the CCPA businesses must provide consumers with a notice ldquoat or before the point of collectionrdquo of personal information which must describe the personal information to be collected and the pur-poses for collecting that information Businesses must additionally allow consumers to request access to and request deletion of personal informa-tion Businesses must allow for consumers to opt-out of the sale of any personal information Developers of COVID-tracing apps would want to build in compliance with these requirements In addition California Civil Code sect 1798815(a)(1) requires companies to ldquomaintain reason-able security procedures and practices appropriate to the nature of the information it processesrdquo
Like privacy laws generally the CCPA does not grant consumers rights regarding the use of de-identified information However the CCPA does require businesses to implement processes that prohibit re-identification of de-identified information as well as technical safeguards to prevent inadvertent release of that information Cal Civ Code sect 1798140(h)
Health Information Privacy
Many DCTT systems will be designed to collect health-related data of users such as symptom tracking SARS-CoV-2 test results and prior ex-posure to a person who is COV+ Individuals may have additional privacy protections with respect to the use and disclosure of this health-related information
The use and disclosure of individually identifiable health information is strictly regulated under the privacy and security rules implementing the
Legal Considerations 83
Health Insurance Portability and Accountability Act (HIPAA) HIPAA is limited in application however to health care providers and health insur-ance plans (ldquocovered entitiesrdquo) and ldquobusiness associatesrdquo of such entities ldquoBusiness associatesrdquo under HIPAA are persons who perform services for covered entities and need access to personal health information to do so
HIPAA-covered entities must have written authorization to use or disclose identifiable health information (ldquoprotected health informationrdquo or PHI) from the individual to whom such information pertains unless the HIPAA regulations promulgated by the US Department of Health and Human Services (HHS) provide an exception to the requirement for such individual authorization
Among the exceptions to the individual authorization requirement is an exception for certain uses and disclosures of PHI for public health purposes 45 CFR sect 164512(b) This exception would permit for exam-ple a HIPAA-covered entity to disclose the PHI of an individual who tests positive for SARS-CoV-2 to a public health authority A ldquopublic health authorityrdquo is an agency or authority of the US government a state ter-ritory a political subdivision of a state or territory or Indian tribe that is responsible for public health matters as part of its official mandate as well as a person or entity acting under a grant of authority from or under a contract with a public health agency such as a contact tracer Id sect 164501
Many DCTT developers are HIPAA business associates and any use and disclosure of PHI collected through DCTT used on behalf of HI-PAA-covered entities is restricted under the HIPAA privacy rules Nota-bly in response to COVID-19 HHS announced that its Office for Civil Rights would exercise its enforcement discretion and would not impose penalties for violations of certain provisions of the HIPAA Privacy Rule against health care providers or their business associates for the good faith uses and disclosures of protected health information for public health and health oversight activities during the nationwide public health emergency 85 FR 19392 (2020)
Many states also have health information privacy laws The HIPAA privacy rule sets a ldquofloorrdquo of privacy protections allowing the states to be more protective of privacy More specifically HIPAA preempts a state law if (but only if) the state law is ldquocontraryrdquo and less protective of privacy than the HIPAA privacy rule However if a state law is determined by the
84 Digital Contact Tracing for Pandemic Response
Secretary of HSS to be necessary to serve a ldquocompelling need related to public health safety or welfarerdquo it may survive preemption even if it is less privacy-protective than HIPAA 45 CFR sect 160203 (a)(1)(iv)
The Public Health Service Act also restricts the use of certain per-sonally identifiable information collected by entities involved with public health activities without the individualrsquos consent 42 USC 242m(d)
Labor and Employment Privacy Rights
Labor and employment lawsmdashthat is laws that govern the relationships between employers and employeesmdashmay prove relevant to DCTT espe-cially if employers mandate the use of DCTT or seek to collect health information regarding their employees using DCTT Depending on the built-in privacy protections of the DCTT system an employer may be able to access important health information from an employeersquos phone As noted above the ECPA would generally prohibit an employer from ac-cessing this information without the employeersquos consent Even with con-sent however there are limits on the collection and use of an employeersquos health information
In particular the use of DCTT may raise special concerns about em-ployment discrimination for example if an employer were to fire an em-ployee who tests positive for SARS-CoV-2 (COV+) or who has a known SARS-CoV-2 exposure The Americans with Disabilities Act (ADA) pro-tects disabled employees from discrimination and restricts the collection of personal health information by employers The Equal Employment Opportunity Commission (EEOC) which is the federal agency tasked with enforcing the ADA in the employment context would likely con-sider COV+ to be a ldquodisabilityrdquo under the ADA and analogous state laws prohibiting discrimination against disabled people COV+ is likely to be a ldquodisabilityrdquo especially where the individual is symptomatic andor experi-ences related health issues or if it is later determined that testing positive for SARS-CoV-2 leads to long-term or chronic health effects ldquoExposure to a COV+ personrdquo could also be covered by those laws because a person exposed to a COV+ individual could well be perceived as being disabled by being considered likely to be infected
Legal Considerations 85
The ADA generally requires that businesses make ldquoreasonable accom-modationsrdquo for persons who are disabled which may include individuals who are COV+ or who have a preexisting disability that places them at higher risk from or may be exacerbated by COVID-19 The EEOC has published guidance on reasonable accommodations under the ADA and related laws in the context of COVID-19 (EEOC 2020) Among other things this guidance clarifies that consistent with the ADA employers may take temperatures or otherwise collect health information about employees during the pandemic crisis so long as they keep that infor-mation confidential As of May 18 2020 the EEOC has not provided guidance that specifically addresses the applicability of the ADA to the use of DCTT by employers
In addition employment laws such as the ADA and the Family and Medical Leave Act (FMLA) and state law equivalents generally limit disclosure of information and require employers to keep confidential any employee personal health information related to a disability or request for medical leave Under the ADA any information regarding the medical condition or history of an employee that an employer obtains as part of an examination or inquiry into a disability could constitute a confidential medical record that can be disclosed only to certain individuals in lim-ited circumstances 42 USC sectsect 12112(d)(3)(B) and 12112(d)(4) The FMLA also prevents the disclosure of records related to medical histories in connection with an employeersquos leave request or eligibility 29 CFR sect 825500(g) The EEOC and some courts have gone further and taken the position that any information concerning an employeersquos medical con-dition is protected under the ADA or FMLA
As discussed elsewhere in this guidance document employers may have a good reason to employ DCTT in order to ensure workplace safety and limit the spread of SARS-CoV-2 in the community Employers may also face legal liability if they fail to protect employees (or customers) from potential exposure or infection In particular employers have an obligation under the Occupational Safety and Health Act to keep the workplace safe for employees In response to COVID-19 the Occupa-tional Safety and Health Administration (OSHA) has developed guid-ance on preparing the workplace (OSHA 2020) The CDC has also pre-pared guidance on healthy business operations and reducing the spread
86 Digital Contact Tracing for Pandemic Response
of SARS-CoV-2 in the workplace (CDC 2020c) Employers must strike an appropriate balance between avoiding employment discrimination and promoting workplace safety
Reflecting the need for such a balance the employee protections un-der the ADA and other employment laws are not absolute and are limited by among other things the need to protect the health and safety of other employees and the public Protection for workplace safety and health generally will justify appropriately tailored measures such as inquiries into an employeersquos personal health status or whether someone has tested positive for SARS-CoV-2 temperature checks and removal of employees from the workplace who are experiencing symptoms or have tested posi-tive and have not been cleared to return to work
Note finally that the use of DCTT by employers should be evaluated in conjunction with the hazard pay sick leave and other benefits that are available to employees Under the Families First Coronavirus Response Act employers with more than 50 employees and fewer than 500 employ-ees are required to provide two weeks of paid sick leave to an employee who stays home because of COVID-19 Pub L No 116-127 134 Stat 178 (2020) This paid leave extends to those who are themselves ill are quarantined or are awaiting a diagnosis as well as those who are caring for sick family members However reporting suggests that more than 75 of US workers will not qualify for benefits under this act (Cochrane Miller and Tankersley 2020)
Constitutional Privacy Rights
A DCTT program involving only private actors operating on the ba-sis of voluntarily provided information would not present constitutional privacy issues But any government-directed use of digital technology to support public health tracking and contact tracing involving mandatory government surveillance may potentially implicate a variety of consti-tutional protections These constitutional protections apply to actions taken by any level of government in the United States While state gov-ernments have broad policing powers in the area of public health (Jacob-son v Massachusetts 197 US 11 (1905)) and are generally allowed to enforce legislation not preempted by federal laws even emergency and
Legal Considerations 87
health-protective laws must be consistent with the US Constitution (HHS 2019 CDC 2020f)
Fourth Amendment Search and Seizure
Many people considering whether to use a DCTT app may be concerned that government enforcement agencies would obtain tracing data and use those data to conduct criminal prosecutions or immigration proceed-ings Constitutional protections notably the Fourth Amendmentrsquos limit on warrantless searches limit the governmentrsquos use of personal data in the criminal context However exceptions exist allowing law enforce-ment to access information even when such access would generally be prohibited How the government accesses personal data stemming from contact tracing needs to be scrutinized and protections will hinge on the manner of access
In general the Fourth Amendment protects ldquo[t]he right of the peo-ple to be secure in their persons houses papers and effects against un-reasonable searches and seizuresrdquo As originally interpreted the Fourth Amendment was considered tied to common-law trespass That is no lon-ger the case US Supreme Court precedent interprets the Fourth Amend-ment to protect ldquopeople not placesrdquo and extends to the protection of certain expectations of privacy such as location information as long as such expectations are reasonable Katz v United States 389 US 347 351 (1967) A warrantless government search is unconstitutional when the information sought is private and such expectation of privacy is ldquoone that society is prepared to recognize as reasonablerdquo Smith v Maryland 442 US 735 743ndash44 (1979)
The constitutionality of a search will revolve around the following analysis whether the digital program either violates an individualrsquos ldquorea-sonable expectation of privacyrdquo (likely triggered by programs collecting large amounts of location andor health data) or involves a government ldquotrespassrdquo (likely triggered by required app downloads) Katz v United States 389 US 347 (1967) United States v Jones 565 US 400 (2012)
Courts will most likely weigh the intrusiveness of the measures taken in implementing a search standard against the severity of the situation governmental and individual interests and accountability measures and safeguards built into the system
Voluntary sharing by individuals of their information with other par-
88 Digital Contact Tracing for Pandemic Response
ties including the government would mean that there was no reasonable expectation of privacy and would not raise the issues elaborated above It is worth noting that consent may not be considered voluntary if coerced or conditioned especially with regard to public employees or students of public institutions
Third-Party Doctrine
Some legal doctrines allow for the governmentrsquos acquisition of otherwise private information consistent with Fourth Amendment privacy protec-tions The third-party doctrine for example provides that individuals have no reasonable expectation of privacy in information voluntarily shared with others even if the information is revealed on the assumption that it will be used only for a limited purpose and the confidence placed in the third party will not be betrayed Smith v Maryland 442 US 735 (1979) United States v Miller 425 US 435 (1976) This applies to in-formation provided by third parties (mobile carriers internet service pro-viders medical tracking device manufacturers etc) to the government under order or request even when the third partyrsquos end-user agreements or privacy policies create an expectation of privacy
The Supreme Court has narrowed the applicability of the third-party doctrine to exclude use and disclosure of ldquohistoricalrdquo cell-site location information (CSLI) data For example in Carpenter v United States 138 S Ct 2206 (2018) the Court reasoned that the third-party doctrine does not justify use and disclosure of historical CSLI because an individual does not provide that information voluntarily Rather that information is pervasively collected by the cell phone company without any affirmative action on the part of the individual The Court did not express a view on ldquoreal-timerdquo CSLImdashlocation information that live-tracks a cell phonersquos locationmdashor on GPS data that may be stored in the phone itself
The Carpenter decision builds on a line of cases related to searches of digitally stored location data In Riley v California 134 S Ct 2473 (2014) the Court held that absent exigent circumstances law enforce-ment must obtain a warrant to search an individualrsquos phone Exigent circumstances are those that require immediate action because there is a probability that evidence may be destroyed The use of a centralized data-base for collection of digital contact tracing data would obviate deletion
Legal Considerations 89
concerns If the data are stored locally in the phone issues may arise as to whether law enforcement may suspect the data may be deleted following an arrest
Similarly in United States v Jones 132 S Ct 945 (2012) Justice So-nia Sotomayor authored a concurring opinion arguing that the use of a GPS to track a defendantrsquos whereabouts has the potential of providing the government with enough data points to create a ldquomosaicrdquo of the personrsquos life Location data obtained through centralized location contact tracing have the potential of providing information on an individualrsquos where-abouts beyond whatrsquos necessary for determining proximity to infected individuals Localized data may also raise the same issues if accessed by law enforcement
Following Carpenter several courts have addressed the constitution-ality of novel location tracking In Massachusetts for instance a federal district court concluded that police use of a ldquopole camerardquo on a utility pole to investigate the movements of an individual constituted a search under the Fourth Amendment United States v Moore-Bush 381 FSupp3d 139 (D Mass 2019) The court reasoned that even in a public space an in-dividual still retains a reasonable expectation of privacy ldquoin the whole of their physical movementsrdquo Citing Carpenter and Jones the court stated that the governmentrsquos unrestrained power to collect data that reveal pri-vate aspects of identity is susceptible to abuse and gives police access to a category of information that is ldquootherwise unknowablerdquo Long-term monitoring of a personrsquos movements consequently violates that individ-ualrsquos expectation of privacy Notably the court emphasized the capability of the camera to create a searchable digital log of the photos taken for the eight-month period during which the camera was used
State courts have also weighed in on the issue The Massachusetts Supreme Judicial Court found that police access to real-time location data pinpointing an individualrsquos movement whether from a third party or a cell-site simulator infringes upon an individualrsquos reasonable expec-tation of privacy Commonwealth v Almonor 120 NE3d 1183 1195 (Mass 2019) The Washington Supreme Court for its part held that a cell phone ping used to locate the defendantrsquos vehicle in real time is a search under the Fourth Amendment requiring a warrant absent exigent circumstances State v Muhammad 428 P3d 1177 (2018) And the Colo-
90 Digital Contact Tracing for Pandemic Response
rado Court of Appeals held that police use of a video pole camera to con-tinuously surveil a defendantrsquos fenced-in backyard constitutes a search under the Fourth Amendment People v Tafoya 2019 BL 457321 Colo Ct App 17CA1243 (2019)
Application of Carpenter by lower courts to novel location-tracking tactics is still evolving and it is as yet unclear how the narrower interpre-tation of the third-party doctrine will continue to be expanded and ap-plied particularly in cases of short-term monitoring of massive amounts of location andor health data Moreover it is unclear whether Carpenter would apply to DCTT data collected by the government itself
Special Needs Doctrine
An argument in favor of the constitutionality of government DCTT programs is that the ldquospecial needsrdquo doctrine would apply Under this doctrine a warrantless search that would otherwise violate the Fourth Amendment might be permissible based on a special need relating to pub-lic health When the search is conducted for a nontraditional law enforce-ment purpose and circumstances make securing a warrant impracticable the Supreme Court has ruled that warrantless searches may be permissi-ble The special needs doctrine however is highly controversial because it is not a consistently applied Fourth Amendment exception so it is diffi-cult to predict when courts would authorize nontraditional surveillance Some factors considered by the court are (1) the balance between the intrusiveness of the government action and the anticipated public bene-fits (2) the existence of legislative authorization (3) judicial process or the ability of the subject individual to challenge the government action (4) the scope or breadth of government action and (5) the likelihood of the collected data being used in criminal proceedings The Supreme Court did note in Chandler v Miller 520 US 305 (1997) that a ldquorisk to public safety [that] is substantial and realrdquo may justify ldquoblanket suspicionless searches calibrated to the riskrdquo citing as examples the routine searches conducted at airports and entrances to some official buildings (Searches within the context of immigration are further analyzed below)
Immigration Enforcement
Exceptions apply to the constitutional requirement that a warrant ac-company an unreasonable search or seizure in the immigration context
Legal Considerations 91
For example an exception to the general warrant requirement is the bor-der search exception which allows government officials to search and seize without a warrant persons and property at the border or at the functional equivalent of a border See United States v Montoya de Her-nandez 473 US 531 (1985) United States v Flores-Montano 541 US 149 (2004) Federal regulation authorizes immigration officials to oper-ate within 100 miles of any US external boundary (See 8 CFR sect 2871 defining ldquoreasonable distancerdquo as ldquowithin 100 air miles from any external boundary of the United Statesrdquo) A functional equivalent of a border may include any airport where international flights may be received automo-bile checkpoints servicing international traffic and vessels in territorial waters Government officials however must still have ldquoreasonable suspi-cionrdquo of an immigration violation or a crime to search or seize persons or property
In the context of digital data Customs and Border Protection (CBP) officials may conduct either manual or forensic searches of electronic devices at the border or its functional equivalent A manual search is considered a routine search and may include accessing the phone and ldquobrowsingrdquo its contents If the electronic device is password protected individuals must provide information for unlocking the device Forensic searches on the other hand are nonroutine and involve a more invasive search of the electronic devicersquos contents Federal circuit courts are split on whether a CBP agent needs ldquoreasonable suspicionrdquo before conducting a forensic search of an electronic device But Supreme Court precedent clearly states that suspicionless searches are not unconstitutional when public safety is considered Skinner v Ry Labor Execsrsquo Assrsquon 489 US 602 (1989)
A recent CBP directive provides guidance and standard operating procedures regarding forensic searches of electronic devices CBP 3340-049A Border Search of Elec Devices (DHS 2018) The directive states that CBP officers may detain electronic devices or copies of the informa-tion contained within these devices for a reasonable period time not to exceed five days This directive raises the concern that travelers may be required to turn over contact tracing data stored on their phone to CBP officers Note that the directive has been challenged in federal court and is currently awaiting appeal Alasaad v Nielsen 419 FSupp3d 142 (D Mass 2019)
92 Digital Contact Tracing for Pandemic Response
Searches in Schools
Another exception to the general warrant requirement applies to searches by non-law-enforcement government officials in public schools (ie school officials) Within this context school officials have broad powers to conduct searches as long as those searches are reasonable Searches by individuals in private schools are not governed by the Fourth Amend-ment State regulation of searches in private schools varies (See US DOE 2009)
Related Federal Privacy Statutes
Outside the Fourth Amendment context certain laws provide protections against government collection of and access to personal data The USA Freedom Act of 2015 for example bans the governmentrsquos bulk collec-tion of internet metadata and telephonic records which was previously allowed under Section 215 of the USA Patriot Act The government must now identify with specificity the identity of a person account address or personal device when requesting records The law allows for the acqui-sition of data by two degrees of separationmdashor ldquohopsrdquomdashfrom targeted individuals If a centralized system in contact tracing is used it is unclear whether the government may need to resort to this provision since it would likely have consent from individuals to collect and use the data
The Privacy Act of 1974 also regulates the collection use and disclo-sure of personal data but applies only to federal agencies (and their con-tractors) not to state or local agencies 5 USC sect 552a The Act protects against disclosure of individually identifying ldquorecord[s]rdquo that are kept within a ldquosystem of recordsrdquo The Act limits disclosure of information ldquoexcept pursuant to a written request by or with prior written consent of the individual to whom the record pertainsrdquo Certain disclosures are ex-empt from the Actrsquos applicability Pertinent disclosure exceptions are for records required to be disclosed under the Freedom of Information Act (FOIA) or disclosures ldquoto a person pursuant to a showing of compelling circumstances affecting the health or safety of an individualrdquo A disclosure under FOIA however would not include information in ldquopersonnel and medical files and similar filesrdquo when disclosure ldquowould constitute a clearly unwarranted invasion of personal privacyrdquo FOIA Guide 2004 Edition
Legal Considerations 93
Exemption 6 If non-anonymized data are turned over to the federal or state governments it is important to consider whether PII would be pro-tected from disclosure under FOIA or state freedom of information laws
Consent
User consent is a cross-cutting issue for evaluating many of the laws and regulations governing personal information privacy discussed in the prior sections In general privacy laws can be justified on the grounds that an individual should have the option to control with various types and de-grees of limitation the collection use retention andor disclosure of in-formation pertaining to that individual by others As such many privacy laws start from the premise that absent an individualrsquos consent use or disclosure of that individualrsquos PII is impermissible except for certain enu-merated purposes deemed to outweigh the individualrsquos privacy interests
Consent like agreements in general can be manifest in different ways in specific circumstances In some cases an affirmative actionmdashsuch as a signaturemdashis needed to demonstrate consent In other cases inactionmdashsuch as declining to ldquounsubscriberdquo from receiving certain unsolicited emailsmdashconstitutes consent Where a law requires a ldquowrittenrdquo signature in all but a few contexts the signature may be executed electronically In the United States that means the ldquosignaturerdquo may consist of any of the following ldquoan electronic sound symbol or processrdquo so long as it is ldquoattached to or logically associated with a contract or other record and executed or adopted by a person with the intent to sign the recordrdquo Electronic Signatures in Global and National Commerce Act (E-SIGN) (15 USC 7006)
The scope of a consent depends on what was deemed to be under-stood by the consenting party That is least clear when the consent is in-ferred from inactivity even if terms stating the consequences of inactivity have been provided The scope of consent is most clear when the terms agreed to have been presented to or provided by the consenting party in a conspicuous documented manner and a record exists of those terms Courts uphold the validity of clickthrough agreements because users are deemed to review the terms to which they respond by clicking ldquoI agreerdquo
94 Digital Contact Tracing for Pandemic Response
But where terms are ambiguous or confusing buried in other text or presented obscurely the ldquoI agreerdquo action may not mean the user actually agreed to specific terms
Terms of Use and Privacy Policies for apps are often written in com-plicated or nuanced language with key points difficult to discern More-over they are generally hard to read on a mobile device Many users of mobile phone apps agree to such terms without even attempting to read or to understand them As such it is often questionable whether an app user has knowingly agreed to all the terms of those documents Presen-tation of terms in large typeface short sentences simple language and direct disclosures makes user consent more meaningful
For contact tracing apps that collect PII andor PHI consent will overcome the restrictions of many if not most privacy laws provided the consent is freely given reflects a full understanding of the terms for use collection and disclosure of the information and is confirmed by an af-firmative act such as a click that may be executed only upon a complete reading of Terms of Use and Privacy Policies Whether consent may be deemed ldquofreely givenrdquo in certain circumstances depends on contextual understandings of party relationships including the employer-employee and government-citizen relationships
Anti-discrimination and Individual Freedom Laws
Any measure taken to protect public health and safety must comply with the Constitution and civil rights laws such as the ADA that prohibit discrimination against persons in certain protected categories such as race gender religion or disability In addition certain implementations of DCTT could be challenged under a variety of individual freedom protections
Anti-discrimination Laws
In general it would be impermissible to use DCTT in a way that either targets or excludes people on the basis of their membership in one of these protected categories
When motivated by animus against a protected class as defined by law and not narrowly tailored to advance a compelling government inter-
Legal Considerations 95
est a discriminatory regulation would be considered unconstitutional un-der the Equal Protection Clause of the Fourteenth Amendment to the US Constitution See Jew Ho v Williamson 103 F10 CCND Cal (1900) (striking down a quarantine imposed by San Francisco in response to an outbreak of bubonic plague because it was racially motivated) see also Church of Lukumi Babalu Aye v Hialeah 508 US 520 (1993) (supposed public health measure unconstitutional because it targeted the practices of one religion)
The risk of unintentional yet illegal discrimination in using DCTT is a real possibility Recent studies of infection rates among the population have revealed a larger-than-proportional infection rate among certain minority communities such as Latinx African American and American Indian communities (NYC DOH 2020) Programs that target specific ra-cial ethnic tribal or religious groups may raise constitutional and other legal concerns
Religious Freedom Laws
The use of DCTT may also raise concerns about religious freedom For example there may be religious objections to restrictions on gathering for worship carrying a mobile phone or the use of imaging technology Under current Supreme Court precedent generally applicable laws that do not discriminate against religion on their face do not violate the Free Exercise Clause of the First Amendment even if those laws have an inci-dental effect on the exercise of religion Employment Div Dept of Hu-man Resources of Oregon v Smith 494 US 872 (1990) These laws need not be justified by compelling government interest (the ldquostrict scrutinyrdquo standard of review) the government need only show that they are ratio-nally related to a legitimate interest On the other hand laws that are not neutral and not of general applicability must be justified by compelling government interest and must be narrowly tailored to advance that in-terest if it burdens religious practicesmdasha very tough hurdle to overcome Church of Lukumi Babalu Aye v Hialeah 508 US 520 (1993)
This general approach however is disrupted in some contexts by statutes adopted to provide greater protection to religious freedom The federal Religious Freedom Restoration Act (RFRA) requires strict scru-tiny for federal actions that burden religion and many states have ad-opted ldquostate RFRAsrdquo that do the same for actions by state and local
96 Digital Contact Tracing for Pandemic Response
governments The Religious Land Use and Institutionalized Persons Act which extends similar protections to persons confined to an institution such as a prison jail or mental health facility may also be relevant 42 USC sect 2000cc
Under either standard of review courts will examine whether a gov-ernment action imposes a substantial burden on religious exercise if not no religious freedom violation has occurred Such a finding is unlikely for DCTT programs absent evidence that the government is using the digital information to take action against religious persons that is not necessary to avoid the spread of a serious disease Nevertheless legal challenges on religious freedom grounds cannot be ruled out
Legislative Recommendations
bull Congress should enact new legislation specifically tailored to fa-cilitate the use of DCTT as part of the public health response to COVID-19 while also protecting user privacy and ensuring data security
bull Congress should require DCTT developers to disclose to users in clear language the nature of the information that would be col-lected how it would be collected how it would be stored and for what purposes it may be used
bull While the rollout of DCTT should initially employ an opt-in au-thorization approach the feasibility acceptability and value of opt-out approaches should continue to be evaluated As such opt-out approaches to consent should not be precluded by legislation
bull Congress should prohibit the commercial use of data collected for COVID-19 response by DCTT
bull Congress should prohibit discrimination on the basis of data col-lected by DCTT
bull If Congress is unable to enact suitable legislation state legislatures should work toward enacting similar laws for their jurisdictions A ldquomodelrdquo state law should be rapidly developed to facilitate na-tionwide uniformity of legal requirements
97
Public Health
Characteristics That Make Data Useful to
Public Health for Reducing Disease Transmission
bull Technologies or apps with the goal of enhancing public health ca-pacity to identify cases and trace contacts in order to control the spread of SARS-CoV-2 should be designed to match functionality with that goal
bull Technologies or apps may produce some false negatives or false positives but they should be accurate enough that public health authorities feel confident that they support and donrsquot detract from contact tracing efforts
bull DCTT approaches for public health should be designed to facili-tate the following
deg Identifying contacts including those who may not be easily found otherwise
deg Finding and notifying contacts rapidly before they develop symptoms if infected
deg Analyzing the nature of contact to determine whether con-tact is high medium or low risk to support decisions about whether quarantine should be mandatory should be volun-tary or is not needed
deg Following up with cases and contacts so that public health can provide resources to support isolation and quarantine at home
Recommendations
FIVE
98 Digital Contact Tracing for Pandemic Response
bull Data collected through DCTT should be made available to public health professionals and to researchers in de-identified form to support population-level epidemiologic analysis
Ethics
Collecting Data to Inform Policy and Practice
bull Reviews of DCTT systems must be conducted in part by an inde-pendent intermediary that has established the publicrsquos trust
bull Those who authorize use of DCTT within a particular jurisdic-tion or institution should continuously and systematically monitor the technologyrsquos performance in that context This should include monitoring for effectiveness and benefit monitoring for harms and monitoring for the fair distribution of both benefits and harms They should also monitor evidence that is being generated in other contexts about their selected technological solution and about other competing technologies
bull Data should be available to users that would permit them to further investigate their personal risk with public health officials or other health workers to add a layer of protection against unnecessary quarantine
Public Trust and Public Attitudes
bull More research into public attitudes is needed In particular in-depth qualitative research should examine public attitudes about perceptions of trust in DCTT among different communities which features of DCTT influence trust and the extent to which people are willing to provide different types of data through DCTT to help their community
bull States and localities that are considering adopting DCTT should engage with the public to increase their understanding of the acceptability of DCTT design features and uses among diverse communities
Recommendations 99
Designing Flexible Technology to Maximize
Public Health Utility While Respecting Other Values
bull Technology companies should not alone control the terms condi-tions or capabilities of DCTT nor should they presume to know what may be acceptable to members of the public
bull A ldquovalues in designrdquo approach to development of DCTT should be adopted (Flanagan Howe and Nissenbaum 2008 Knobel and Bowker 2011) Robust public- and user-engagement activities should be pursued to identify and incorporate to the extent pos-sible a range of values into the design of the technology These values may include privacy but also autonomy efficiency equity or others Technology design should reflect an appropriate balance and prioritization of identified values
bull Technology design should not be static but should be capable of evolving depending upon local conditions new evidence and changing preferences and priorities
bull DCTT should be designed to have a base set of features that pro-tect privacy with layers of additional capabilities that users may choose to activate An initial default should be that user location data are not shared but users should be provided with easy mech-anisms and prompts to allow for opting-in to this capability with encouragement to the public if and as it is shown to be critical to achieving public health goals
Policy Positions to Advance Widespread
Use of Digital Contact Tracing Technologies
bull DCTT use should not be mandated at this time given uncertainty about potential harms and benefits Additional technology user and real-world testing is needed
bull Incentives can be a useful complement to encouragements how-ever any incentives for users to install and use DCTT must be equitable should not be coercive and should align with effective use of the technology
100 Digital Contact Tracing for Pandemic Response
bull Trusted leaders should be enlisted to communicate effectively with the public about DCTT and encourage its use should the technol-ogy demonstrate some potential The limits of knowledge regard-ing effectiveness should also be explained along with what will be done to improve technological capabilities as understanding evolves
Disclosure and AuthorizationConsent
bull A clear and concise module consisting of basic disclosure and vol-untary authorization should be developed to accompany DCTT This module should not take the form of ldquoclickwraprdquo terms of service or end-user agreements but rather provide only essential information necessary for an individual to make a decision More detailed disclosures (such as FAQs in plain language) should be made easily accessible to those who wish to learn more with no hidden surprises
bull An opt-in approach to authorization should be instituted to ac-company initial DCTT rollout The feasibility and value of opt-out approaches should continue to be evaluated informed by what is technologically possible what local assessments of benefits and harms of the technology reveal over time and our evolving un-derstanding of the degree to which an opt-out approach is likely to increase or decrease utilization among different populations Opt-out approaches should not be precluded
Promoting Equity and Fairness
bull A commitment to equity means a commitment to ensuring that the benefits and burdens of DCTT are distributed fairly Public engagement is an important tool for assessing impact and to rectify inequities
bull States localities and institutions that recommend widespread use of DCTT should provide technology (eg mobile phones Blue-tooth devices) and free data packages to those who desire but lack access to these devices
Recommendations 101
bull If there are lower rates of adoption of DCTT systems in some identifiable communities public health authorities should iden-tify ways to compensate For example directing more non-DCTT resources and efforts toward those communities to meet specific needs that are elsewhere being supported by technology
bull If maps are generated based on DCTT to provide the public with the locations that COV+ individuals have visited steps must be taken to minimize the stigma and potential financial losses that could result from being identified as a hotspot
Instituting Transparent Governance and Oversight
bull Digital surveillance oversight committees should be established expeditiously with diverse and qualified membership to provide ethical and regulatory review prior to and concurrent with wide-spread use of a DCTT system
bull Understandable and publicly accessible rules must guide the col-lection access control use storage and combination of data by government authorities public and private institutions and other parties such as public health researchers
bull Only those data that are necessary and relevant for the public health response to COVID-19 should be collected and used
bull Identifiable data should be kept only for the period of time needed for the public health response to COVID-19
bull Identifiable data collected as part of this response should not be shared with anyone other than the relevant public health authorities without additional specific informed consent of individual users
bull Before a government or institution adopts a digital contact tracing program they should state the conditions under which the digital contact tracing program will be terminated
bull Future use of DCTT to advance public health or other efforts (eg use in seasonal flu surveillance) would require independent justification DCTT designed for public health use should not be used by law or immigration enforcement
102 Digital Contact Tracing for Pandemic Response
bull The principles offered in this guidance document apply both during and following the COVID-19 pandemic
Legislative
bull Congress should enact new legislation specifically tailored to fa-cilitate the use of DCTT as part of the public health response to COVID-19 while also protecting user privacy and ensuring data security
bull Congress should require DCTT developers to disclose to users in clear language the nature of the information that would be col-lected how it would be collected how it would be stored and for what purposes it may be used
bull While the rollout of DCTT should initially employ an opt-in au-thorization approach the feasibility acceptability and value of opt-out approaches should continue to be evaluated As such opt-out approaches to consent should not be precluded by legislation
bull Congress should prohibit the commercial use of data collected for COVID-19 response by DCTT
bull Congress should prohibit discrimination on the basis of data col-lected by DCTT
bull If Congress is unable to enact suitable legislation state legislatures should work toward enacting similar laws for their jurisdictions A ldquomodelrdquo state law should be rapidly developed to facilitate na-tionwide uniformity of legal requirements
Resources
103
US Government Response
White House
The White House and CDC 16 April 2020 ldquoOpening Up America Againrdquo The White House and the Centers for Disease Control and Prevention Available at httpswwwwhitehousegovopeningamerica
Congress
A Bill to Protect the Privacy of Consumersrsquo Personal Health Information
Proximity Data Device Data and Geolocation Data during the Corona-
virus Public Health Crisis S3663 116th Cong (2019-2020) Available at httpswwwcongressgovbill116th-congresssenate-bill3663
A Bill to Protect the Privacy of Health Information during a National
Health Emergency S3749 116th Cong (2019-2020) Available at httpswwwcongressgovbill116th-congresssenate-bill3749
To Protect the Privacy of Health Information during a National Health
Emergency HR6866 116th Cong (2019-2020) Available at httpswwwcongressgovbill116th-congresshouse-bill6866
Congressional Research Service
Foster Michael 16 April 2020 ldquoCOVID-19 Digital Surveillance and Privacy Fourth Amendment Considerationsrdquo Legal Sidebar LSB10449 Congres-sional Research Service httpscrsreportscongressgovproductpdfLSBLSB10449
Mulligan Stephen P and Chris D Linebaugh 25 March 2019 ldquoData Pro-tection Law An Overviewrdquo R45631 Congressional Research Service httpscrsreportscongressgovproductpdfRR45631
104 Resources
US Department of Health and Human Services (HHS) Centers for Disease Control and Prevention (CDC)
CDC 6 May 2020 ldquoInterim Guidance for Businesses and Employersrdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention Available at httpswwwcdcgovcoronavirus2019-ncov communityguidance-business-responsehtml
CDC 3 May 2020 ldquoDiscontinuation of Isolation for Persons with COVID-19 Not in Healthcare Settingsrdquo Interim Guidance Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention httpswwwcdc govcoronavirus2019-ncovhcpdisposition-in-home-patientshtml
CDC 30 April 2020 ldquoContact Tracingrdquo Get and Keep America Open Sup-porting States Tribes Localities and Territories Centers for Disease Control and Prevention httpswwwcdcgovcoronavirus2019-ncovphpopen-americacontact-tracinghtml
CDC 29 April 2020 ldquoInterim Guidelines for Collecting Handling and Testing Clinical Specimens from Persons for Coronavirus Disease 2019 (COVID-19)rdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention Available at httpswwwcdcgovcorona-virus2019-ncovlabguidelines-clinical-specimenshtml
CDC 29 April 2020 ldquoContact Tracing Part of a Multipronged Approach to Fight the COVID-19 Pandemicrdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention httpswwwcdcgov coronavirus2019-ncovphpprinciples-contact-tracinghtml
CDC 28 April 2020 ldquoPreliminary Criteria for the Evaluation of Digital Contact Tracing Tools for COVID-19rdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention Available at httpswwwcdc govcoronavirus2019-ncovdownloadsphpprelim-eval-criteria-digital- contact-tracingpdf
CDC 20 April 2020 ldquoDigital Contact Tracing Tools for COVID-19rdquo Corona- virus Disease 2019 (COVID-19) Centers for Disease Control and Preven-tion Available at httpswwwcdcgovcoronavirus2019-ncovdownloadsdigital-contact-tracingpdf
CDC 6 April 2020 ldquoOperational Considerations for the Identification of Healthcare Workers and Inpatients with Suspected COVID-19 in Non-US Healthcare Settingsrdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention Available at httpswwwcdcgovcoron virus2019-ncovhcpnon-us-settingsguidance-identify-hcw-patientshtml
HHS 2 April 2020 ldquoNotification of Enforcement Discretion under HIPAA to Allow Uses and Disclosures of Protected Health Information by Business
Resources 105
Associates for Public Health and Health Oversight Activities in Response to COVID-19rdquo 45 CFR Parts 160 and 164 Available at httpswwwhhsgovsitesdefaultfilesnotification-enforcement-discretion-hipaapdf
HHS 30 March 2020 Notification of Enforcement Discretion for Telehealth Remote Communications during the COVID-19 Nationwide Public Health Emergency US Department of Health and Human Services Office of Civil Rights Available at httpswwwhhsgovhipaafor-professionalsspecial- topicsemergency-preparednessnotification-enforcement-discretion- telehealthindexhtml
CDC 24 February 2020 ldquoLegal Authorities for Isolation and Quarantinerdquo Centers for Disease Control and Prevention Available at httpswww cdcgovquarantineaboutlawsregulationsquarantineisolationhtml
HHS 26 November 2019 ldquoPublic Health Emergency Declarationrdquo Public Health Emergency US Department of Health and Human Services Office of the Assistant Secretary for Preparedness and Response Available at httpswwwphegovPreparednesslegalPagesphedeclarationaspx
CDC September 2018 ldquoPublic Health Surveillance Preparing for the Futurerdquo Centers for Disease Control and Prevention Available at httpswww cdcgovsurveillancepdfsSurveillance-Series-Booklethpdf
CDC 2005 ldquoVI Plan for Surveillance of Contacts of SARS Case Supplement B SARS Surveillancerdquo Centers for Disease Control and Prevention Available at httpswwwcdcgovsarsguidanceb-surveillancecontactshtml
National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research 1979 ldquoThe Belmont Report Ethical Principles and Guidelines for the Protection of Human Subjects of Researchrdquo US Depart-ment of Health and Human Services Office of Human Research Protec-tions httpswwwhhsgovohrpregulations-and-policybelmont-reportread-the-belmont-reportindexhtml
US Equal Employment Opportunity Commission (EEOC)
EEOC 7 May 2020 ldquoWhat You Should Know About COVID-19 and the ADA the Rehabilitation Act and Other EEO Lawsrdquo US Equal Employ-ment Opportunity Commission Available at httpswwweeocgovwyskwhat-you-should-know-about-covid-19-and-ada-rehabilitation-act-and-other-eeo-laws
EEOC 21 March 2020 ldquoPandemic Preparedness in the Workplace and the Americans with Disabilities Actrdquo US Equal Employment Oppor-tunity Commission Available at httpswwweeocgovlawsguidancepandemic-preparedness-workplace-and-americans-disabilities-act
106 Resources
Federal Trade Commission (FTC)
FTC 2020 ldquoPrivacy amp Data Security Update 2019rdquo Federal Trade Commis-sion httpswwwftcgovsystemfilesdocumentsreportsprivacy-data- security-update-20192019-privacy-data-security-report-508pdf
FTC 2012 ldquoProtecting Consumer Privacy in an Era of Rapid Changerdquo FTC Report Recommendations for Businesses and Policymakers Fed-eral Trade Commission httpswwwftcgovsitesdefaultfilesdocumentsreportsfederal-trade-commission-report-protecting-consumer-privacy-era- rapid-change-recommendations120326privacyreportpdf
Occupational Safety and Health Administration (OSHA)
OSHA April 2020 ldquoGuidance on Preparing Workplaces for COVID-19rdquo OSHA 3990-03 2020 Department of Labor Occupational Safety and Health Act of 1970 Available at httpswwwoshagovPublicationsOSHA3990pdf
Customs and Border Protection (CBP)
CBP 4 January 2018 ldquoBorder Search of Electronic Mediardquo US Customs and Border Protection CDP DIRECTIVE NO 3340-049A Available at httpswwwcbpgovsitesdefaultfilesassetsdocuments2018-JanCBP-Directive-3340-049A-Border-Search-of-Electronic-Media-Compliantpdf
Other Governmental and Nongovernmental Organizations
Contact Tracing Surveillance PlansMethods
Simmons-Duffin Selena 7 May 2020 ldquoStates Nearly Doubled Plans for Contact Tracers since NPR Surveyed Them 10 Days Agordquo NPR Available at httpswwwnprorgsectionshealth-shots20200428846736937we-asked-all-50-states-about-their-contact-tracing-capacity-heres-what- we-learned
Resolve to Save Lives 2020 ldquoCOVID-19 Contact Tracing Playbookrdquo Vital Strategies Available at httpscontacttracingplaybookresolvetosavelives org
Simpson Erin and Adam Conner 22 April 2020 ldquoDigital Contact Tracing to Contain the Coronavirusrdquo Center for American Progress Avail-able at httpswwwamericanprogressorgissuestechnology-policynews20200422483521digital-contact-tracing-contain-coronavirus
Watson Crystal Anita Cicero James Blumenstock and Michael Fraser 10 April 2020 ldquoA National Plan to Enable Comprehensive COVID-19 Case Finding and Contact Tracing in the USrdquo Johns Hopkins Bloomberg School of Pub-
Resources 107
lic Health Center for Health Security and the Association of State and Ter-ritorial Health Officials Available at httpswwwcenterforhealthsecurity orgour-workpubs_archivepubs-pdfs2020200410-national-plan-to- contact-tracingpdf
Africa CDC 9 April 2020 ldquoGuidance on Contact Tracing for COVID-19 Pandemicrdquo Manuals Guidelines amp Frameworks African Union Africa CDC Available at httpsafricacdcorgdownloadguidance-on-contact- tracing-for-covid-19-pandemic
PIH 4 April 2020 ldquoPart I Testing Contact Tracing and Community Manage-ment of COVID-19rdquo PIH Guide | COVID-19 STOP COVID Partners in Health Available at httpswwwpihorgsitesdefaultfiles2020-04PIH_Guide_COVID_Part_I_Testing_Tracing_Community_Managment_4_4pdf
Hellewell Joel Sam Abbott Amy Gimma et al 28 February 2020 ldquoFea- sibility of Controlling COVID-19 Outbreaks by Isolation of Cases and Contactsrdquo The Lancet Global Health 8 e488ndash96 Available at https doiorg101016S2214-109X(20)30074-7
Contact Tracing EthicsPrinciplesGuidance
Center for Democracy amp Technology (CDT)ndashhttpscdtorginsightsCDT 30 April 2020 ldquoStatement of the Center for Democracy amp Technol-
ogy Regarding Use of Data to Fight COVID-19rdquo Center for Democ-racy amp Technology httpscdtorgwp-contentuploads202004 CDT-Statement-Regarding-Use-of-Data-to-Fight-COVID-19pdf
Shetty Ridhi 23 April 2020 ldquoData Use in the Fight against COVID-19 Should Treat People Equitably Not Exacerbate Long-Standing Dispar-itiesrdquo Center for Democracy amp Technology httpscdtorginsightsdata-use-in-the-fight-against-covid-19-should-treat-people-equitably-not-exacerbate-long-standing-disparities
Greenwood Dazza Gregory Nadeau Pagona Tsormpatzoudi Bryan Wilson Jeffrey Saviano and Alex ldquoSandyrdquo Pentland 30 April 2020 ldquoCOVID-19 Contact Tracing Privacy Principlesrdquo MIT Computational Law Report Available at httpslawmitedupubcovid19contacttracingprivacy- principlesrelease7
Editorial Board 29 April 2020 ldquoShow Evidence That Apps for COVID-19 Contact-Tracing Are Secure and Effectiverdquo Nature 580 (7805) 563ndash563 httpsdoiorg101038d41586-020-01264-1
ldquoJoint Statement on Contact Tracingrdquo 19 April 2020 Available at httpscryptobriefingcomwp-contentuploads202004Joint-State-ment-from-Researcherspdf
Hinch Robert Will Probert Anel Nurtay Michelle Kendall Chris Wymant
108 Resources
Matthew Hall Katrina Lythgoe et al 16 April 2020 ldquoEffective Configura-tions of a Digital Contact Tracing App A Report to NHSXrdquo Available at httpsgithubcomBDI-pathogenscovid-19_instant_tracingblobmasterReportndashEffective Configurations of a Digital Contact Tracing Apppdf
Kahn Gilmor Daniel 16 April 2020 ldquoPrinciples for Technology-Assisted Contact-Tracingrdquo White Paper American Civil Liberties Union httpswwwacluorgreportaclu-white-paper-principles-technology-assisted- contact-tracing
Carroll Anna and Samantha Stroman 16 April 2020 ldquoFind My Friends in a Pandemic The Future of Contact Tracing in Americardquo CSIS Commis-sion on Strengthening Americarsquos Health Security April 16 2020 httpshealthsecuritycsisorgarticlesfind-my-friends-in-a-pandemic-the-future- of-contact-tracing-in-america
Electronic Privacy Information Center Testimony to Congress 15 April 2020 ldquoEPIC to Congress Establish Privacy Safeguards for Digital Contact Tracingrdquo httpsepicorgtestimonycongressEPIC-HEC-Contact-Tracing-Apr2020pdf
Crocker Andrew Kurt Opsahl and Bennett Cyphers 10 April 2020 ldquoThe Chal-lenge of Proximity Apps for COVID-19 Contact Tracingrdquo Electronic Fron-tier Foundation April 10 2020 httpswwwefforgdeeplinks202004challenge-proximity-apps-covid-19-contact-tracing
Hart Vi Divya Siddarth Bethan Cantrell Lila Tretikov Peter Eckersley John Langford Scott Leibrand et al 3 April 2020 ldquoOutpacing the Virus Digital Response to Containing the Spread of COVID-19 While Mitigat-ing Privacy Risksrdquo Whitepaper 5 COVID-19 Rapid Response Impact Initiative Edmond J Safra Center for Ethics httpsdrivegooglecomfiled1vIN2AX-DDNW-S0aHq8xs0RJ2jkR_CckXview
Raskar Ramesh Isabel Schunemann Rachel Barbar Kristen Vilcans Jim Gray Praneeth Vepakomma Suraj Kapa Andrea Nuzzo Rajiv Gupta et al 19 March 2020 ldquoApps Gone Rogue Maintaining Personal Privacy in an Epidemicrdquo White Paper Private Kit MIT httpsarxivorgpdf2003 08567pdf
General (not COVID-19 specific) Statements of Principles Legal Frameworks Other Information
Nuffield Council on Bioethics 2020 ldquoGuide to the Ethics of Surveillance and Quarantine for Novel Coronavirusrdquo httpswwwnuffieldbioethicsorgassetspdfsGuide-to-the-ethics-of-surveillance-and-quarantine-for-novel-coronaviruspdf
Resources 109
Schwartz Adam 2020 ldquoHow EFF Evaluates Government Demands for New Surveillance Powersrdquo Electronic Frontier Foundation April 3 2020 httpswwwefforgdeeplinks202004how-eff-evaluates-government-demands- new-surveillance-powers
WHO 2 April 2020 ldquoCoronavirus Disease 2019 (COVID-19)rdquo Situation Report 73 World Health Organization httpswwwwhointdocsdefault-sourcecoronavirusesituation-reports20200402-sitrep-73-covid-19pdf
Nuffield Council on Bioethics 17 March 2020 ldquoEthical Considerations in Responding to the COVID-19 Pandemicrdquo Rapid Policy Briefing Nuffield Council on Bioethics httpswwwnuffieldbioethicsorgassetspdfs Ethical-considerations-in-responding-to-the-COVID-19-pandemicpdf
Nuffield Council on Bioethics 28 January 2020 ldquoResearch in Global Health Emergencies Ethical Issuesrdquo httpswwwnuffieldbioethicsorgassetspdfsRGHE_full_report1pdf
Vota Wayan 4 December 2019 ldquoEvery African Countryrsquos National eHealth Strategy or Digital Health Policyrdquo ICT Works httpswwwictworksorgafrican-national-ehealth-strategy-policy
ENISA 3 December 2019 ldquoPseudonymisation Techniques and Best Practicesrdquo European Union Agency for Cybersecurity httpswwwenisaeuropaeupublicationspseudonymisation-techniques-and-best-practices
ldquoMobile Cellular Subscriptions (per 100 People)rdquo 2018 World Telecommuni-cationICT Development Report International Telecommunication Union httpsdataworldbankorgindicatorITCELSETSP2
Levinson-Waldman Rachel 2018 ldquoCellphones Law Enforcement and the Right to Privacyrdquo Brennan Center for Justice New York University School of Law httpswwwbrennancenterorgsitesdefaultfiles2019-08Report_Cell_Surveillance_Privacypdf
Code of Federal Regulations Part 46ndashProtection of Human Subjects (45 CFR 46 US Department of Health and Human Services)
General Data Protection Regulation (REGULATION (EU) 2016679 (GDPR) European Union)
European Data Protection Supervisor ldquoNecessity and Proportionalityrdquo httpsedpseuropaeudata-protectionour-worksubjectsnecessity- proportionality_en
ISOIEC 38505-12017 Information technologymdashGovernance of ITmdashGover-nance of datamdashPart 1 Application of ISOIEC 38500 to the governance of data (2017 International Organization for Standardization (ISO))
WHO 2016 ldquoWHO Guidance for Managing Ethical Issues in Infectious Dis-ease Outbreaksrdquo World Health Organization httpswwwwhointethicspublicationsinfectious-disease-outbreaksen
110 Resources
CIOMS 2016 ldquoInternational Ethical Guidelines for Health-Related Research Involving HumansndashGuideline 22 Use of Data Obtained from the Online Environment and Digital Tools in Health Related Researchrdquo Council for International Organizations of Medical Sciences httpsciomschwp- contentuploads201701WEB-CIOMS-EthicalGuidelinespdf
Principles for Digital Development 2016 ldquoPrinciplesrdquo httpsdigitalprinciplesorgprinciples
GA4GH 2014 ldquoFramework for Responsible Sharing of Genomic and Health-Related Datardquo Global Alliance for Genomics amp Health 9 Decem-ber 2014 httpswwwga4ghorggenomic-data-toolkitregulatory-ethics- toolkitframework-for-responsible-sharing-of-genomic-and-health-related- data
WHO and PATH 2013 ldquoPlanning an Information Systems Project A Toolkit for Public Health Managersrdquo World Health Organization amp PATH httpspathazureedgenetmediadocumentsTS_opt_ict_toolkitpdf
ASTHO 2012 ldquoPublic Health Collection Use Sharing and Protection of Informationrdquo Issue Brief ASTHO Legal Preparedness Series Association of State and Territorial Health Officials httpswwwasthoorgProgramsPreparednessPublic-Health-Emergency-LawPublic-Health-and-Informat- ion-Sharing-ToolkitCollection-Use-Sharing-and-Protection-Issue-Brief
Cavoukian Ann 2010 ldquoPrivacy by Design The 7 Foundational Principlesrdquo Ontario Canada Information and Privacy Commissioner of Canada httpsiapporgmediapdfresource_centerpbd_implement_7found_ principlespdf
PHLS 2002 ldquoPrinciples of the Ethical Practice of Public Healthrdquo Public Health Leadership Society httpswwwaphaorg-mediafilespdfmembergroupsethicsethics_brochureashx
National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research 1979 ldquoThe Belmont Report Ethical Principles and Guidelines for the Protection of Human Subjects of Researchrdquo US Dept of Health amp Human Services Office of Human Research Protections httpswwwhhsgovohrpregulations-and-policybelmont-reportread-the-bel mont-reportindexhtml
Reopening ProposalsPlans (Broad)
United Kingdom Cabinet Office 2020 ldquoOur Plan to Rebuild The UK Govern-mentrsquos COVID-19 Recovery Strategyrdquo 12 May 2020 httpswwwgovukgovernmentpublicationsour-plan-to-rebuild-the-uk-governments-covid- 19-recovery-strategyour-plan-to-rebuild-the-uk-governments-covid-19- recovery-strategyfourteen-supporting-programmes
Resources 111
Shannon Joel Lorenzo Reyes and Doyle Rice 2020 ldquoAre Lockdowns Being Relaxed in My State Herersquos How America Is Reopening amid the Coronavirus Pandemicrdquo USA TODAY May 21 2020 httpswww usatodaycomstorynewshealth20200419coronavirus-lockdown- reopening-states-us-texas-florida5155269002
Romer Paul 2020 ldquoRoadmap to Responsibly Reopen Americardquo 23 April 2020 httpsroadmappaulromernetpaulromer-roadmap-reportpdf
Governor Larry Hogan 24 April 2020 ldquoMaryland Strong Roadmap to Recoveryrdquo httpsgovernormarylandgovwp-contentuploads202004MD_Strongpdf
NGA and ASTHO 21 April 2020 ldquoRoadmap to Recovery A Public Health Guide for Governorsrdquo National Governors Association and American State and Territorial Health Officials httpswwwngaorgwp-contentuploads202004NGA-Reportpdf
Allen Danielle Julius Krein Ganesh Sitaraman and E Glen Weyl 2020 ldquoNational Covid-19 Testing Action Planrdquo The Rockefeller Foundation httpswwwrockefellerfoundationorgwp-contentuploads202004TheRockefellerFoundation_WhitePaper_Covid19_4_22_2020pdf
Edmond J Safra Center for Ethics Harvard UniverityndashWhite PapersAllen Danielle Sharon Block Joshua Cohen et al 20 April 2020 ldquoRoad-
map to Pandemic Resiliencerdquo httpsethicsharvardedufilescenter- for-ethicsfilesroadmaptopandemicresilience_updated_42020_0pdf
Allen Danielle Lucas Stanczyk Rajiv Sethi Glen Weyl 25 March 2020 ldquoWhen Can We Go Outrdquo httpsdrivegooglecomfiled1gf21eYeNWwrR9OO5nzxn1jlv-RTmHkt0view
Mulheirn Ian Sam Alvis Lizzie Insall James Browne Christina Palmou 20 April 2020 ldquoA Sustainable Exit Strategy Managing Uncertainty Minimis-ing Harmrdquo Tony Blair Institute for Global Change httpsinstituteglobalsitesdefaultfilesinline-filesA20Sustainable20Exit20Strategy2C 20Managing20Uncertainty20Minimising20Harmpdf
Emanuel Zeke Neera Tanden Adam Conner Erin Simpson Nicole Rap-fogel and Maura Calsyn 2020 ldquoA National and State Plan to End the Coronavirus Crisisrdquo Center for American Progress April 3 2020 httpswwwamericanprogressorgissueshealthcarenews20200403482613national-state-plan-end-coronavirus-crisis
Gottlieb Scott Caitlin Rivers Mark McClellan Lauren Silvis and Crystal Watson 2020 ldquoNational Coronavirus Response A Road Map to Reopen-ingrdquo American Enterprise Institute httpswwwaeiorgresearch-productsreportnational-coronavirus-response-a-road-map-to-reopening
112 Resources
Digital Contact Tracing Experiences from Other Countries
Multiple
Woodhams Samuel 12 May 2020 ldquoCOVID-19 Digital Rights Trackerrdquo Top10VPN httpswwwtop10vpncomresearchinvestigationscovid-19- digital-rights-tracker
Fahim Kareem Min Joo Kim and Steve Hendrix 2 May 2020 ldquoCellphone Monitoring Is Spreading with the Coronavirus So Is an Uneasy Tolerance of Surveillancerdquo Washington Post (Washington DC) httpswww washingtonpostcomworldcellphone-monitoring-is-spreading-with-the- coronavirus-so-is-an-uneasy-tolerance-of-surveillance20200502 56f14466-7b55-11ea-a311-adb1344719a9_storyhtml
Ikram Umar Christer Mjaringset MD Anne-Marie Boxall Mylaine Breton Ines Gravey Holly Krelle Veacuteronique Raimond and Reginald D Williams II 30 April 2020 ldquoWhat Can the US Learn from Innovative Strategies Used in Other Countries to Respond to COVID-19rdquo The Commonwealth Fund httpswwwcommonwealthfundorgblog2020what-can-us-learn-innovative-strategies-used-other-countries-respond-covid-19
Jens-Henrik Jeppesen and Pasquale Esposito 29 April 2020 ldquoCOVID-19 European Data Collection and Contact Tracing Measuresrdquo Center for Democracy amp Technology httpscdtorginsightscovid-19-european- data-collection-and-contact-tracing-measures
Africa CDC 9 April 2020 ldquoGuidance on Contact Tracing for COVID-19 Pan-demicrdquo Manuals Guidelines amp Frameworks African Union Africa CDC httpsafricacdcorgdownloadguidance-on-contact-tracing-for-covid-19- pandemic
Heneghan Carl Jon Brassey and Tom Jefferson 6 April 2020 ldquoCOVID-19 What Proportion Are Asymptomaticrdquo Centre for Evidence-Based Medi-cine httpswwwcebmnetcovid-19covid-19-what-proportion-are- asymptomatic
Kharpal Arjun 30 March 2020 ldquoUse of Surveillance to Fight Coronavirus Raises Concerns about Government Power after Pandemic Endsrdquo CNBC httpswwwcnbccom20200327coronavirus-surveillance-used-by- governments-to-fight-pandemic-privacy-concernshtml
Canada
ldquoCommissioner Publishes Framework to Assess Privacy-Impactful Initiatives in Response to COVID19rdquo Office of the Privacy Commissioner of Canada 17 April 2020 httpswwwprivgccaenopc-newsnews-and-announcements 2020an_200417
Resources 113
China
Kraemer Moritz U G Chia-Hung Yang Bernardo Gutierrez Chieh-Hsi Wu Brennan Klein David M Pigott Open COVID-19 Data Working Group Louis du Plessis Nuno R Faria Ruoran Li William P Hanage John S Brownstein Maylis Layan Alessandro Vespignani Huaiyu Tian Chris-topher Dye Oliver G Pybus Samuel V Scarpino ldquoThe Effect of Human Mobility and Control Measures on the COVID-19 Epidemic in Chinardquo Science 368(6490) 493-497 DOI 101126scienceabb4218
Bi Qifang Yongsheng Wu Shujiang Mei Chenfei Ye Xuan Zou Zhen Zhang Xiaojian Liu Lan Wei Shaun A Truelove Tong Zhang Wei Gao Cong Cheng Xiujuan Tang Xiaoliang Wu Yu Wu Binbin Sun Suli Huang Yu Sun Juncen Zhang Ting Ma Justin Lessler and Teijian Feng ldquoEpidemiol-ogy and Transmission of COVID-19 in 391 Cases and 1286 of Their Close Contacts in Shenzhen China A Retrospective Cohort Studyrdquo The Lancet April 27 2020 DOIhttpsdoiorg101016S1473-3099(20)30287-5
Sun Kaiyuan and Ceacutecile Viboud ldquoImpact of contact tracing on SARS-CoV-2 transmissionrdquo The Lancet April 27 2020 DOIhttpsdoiorg101016S1473-3099(20)30357-1
Ferretti Luca Chris Wymant Michelle Kendall Lele Zhao Anel Nurtay Lucie Abeler-Doumlrner Michael Parker David Bonsall and Christophe Fraser 2020 ldquoQuantifying SARS-CoV-2 Transmission Suggests Epidemic Control with Digital Contact Tracingrdquo Science 368 (6491) httpsdoiorg101126scienceabb6936
Gan Nectar and David Culver 2020 ldquoChina Is Fighting the Coronavirus with a Digital QR code Herersquos How It Worksrdquo CNN Business April 16 2020 httpswwwcnncom20200415asiachina-coronavirus-qr-code-intl-hnkindexhtml
Mozur Paul Raymond Zhong and Aaron Krolik 2020 ldquoIn Coronavirus Fight China Gives Citizens a Color Code with Red Flagsrdquo New York Times (New York NY) March 1 2020 httpswwwnytimescom20200301businesschina-coronavirus-surveillancehtml
Germany
Schwartz Matthew S 2020 ldquoGermany Backs Away from Compiling Coro-navirus Contacts in a Central Databaserdquo NPR April 27 2020 https wwwnprorgsectionscoronavirus-live-updates20200427846046185 germany-backs-away-from-compiling-coronavirus-contacts-in-a-central- databaseutm_medium=RSSamputm_campaign=news
Busvine Douglas and Andreas Rinke 2020 ldquoGermany Flips to Apple-Google
114 Resources
Approach on Smartphone Contact Tracingrdquo Reuters April 26 2020 httpswwwreuterscomarticleus-health-coronavirus-europe-techgermany-flips-on-smartphone-contact-tracing-backs-apple-and-google-idUSKCN22807J
Hong Kong
ldquo lsquoStayHomeSafersquo Mobile App User Guiderdquo The Government of the Hong Kong
Special Administrative Region May 20 2020httpswwwcoronavirusgov hkengstay-home-safehtml
India
Dixit Pranav 2020 ldquoIndiarsquos Contact Tracing App Is All But Mandatory So This Programmer Hacked It So That He Always Appears Saferdquo BuzzFeed News May 12 2020 httpswwwbuzzfeednewscomarticlepranavdixitindia-aarogya-setu-hacked
Greenberg Andy 2020 ldquoIndiarsquos Covid-19 Contact Tracing App Could Leak Patient Locationsrdquo WIRED May 6 2020 httpswwwwiredcomstoryindia-covid-19-contract-tracing-app-patient-location-privacy
Alderson Elliot ldquoAarogya Setu The Story of a Failurerdquo Medium May 6 2020 httpsmediumcomfs0c131yaarogya-setu-the-story-of-a-failure- 3a190a18e34
OrsquoNeill Patrick Howell ldquoIndia Is Forcing People to Use Its COVID App Unlike Any Other Democracyrdquo MIT Technology Review May 6 2020 httpswwwtechnologyreviewcom202005071001360india-aarogya- setu-covid-app-mandatory
Israel
Hendrix Steve and Ruth Eglash 2020 ldquoIsrael Is Using Cellphone Surveil-lance to Warn Citizens You May Already Be Infectedrdquo Washington Post (Washington DC) March 19 2020 httpswwwwashingtonpostcomworldmiddle_eastisrael-is-using-cellphone-surveillance-to-warn-citizens-you-may-already-be-infected2020031968267294-69e7-11ea-b199-3a9799c54512_storyhtml
Singapore
ldquoTraceTogether Safer Togetherrdquo A Singapore Government Agency Website accessed May 21 2020 httpswwwtracetogethergovsg
Resources 115
South Korea
Korean Ministry of Health and Welfare ldquoConfirmed Patient Movement Path Websiterdquo Central Accident Remediation Headquarters accessed May 21 2020 httpncovmohwgokrbdBoardList_RealdobrdId=1ampbrdGu-bun=12ampncvContSeq=ampcontSeq=ampboard_id=ampgubun=
Kim Max S 2020 ldquoSeoulrsquos Radical Experiment in Digital Contact Tracingrdquo The New Yorker April 17 2020 httpswwwnewyorkercomnewsnews-deskseouls-radical-experiment-in-digital-contact-tracing
Watson Ivan and Sophie Jeong 2020 ldquoCoronavirus Mobile Apps Are Surging in Popularity in South Koreardquo CNN Business February 28 2020 httpseditioncnncom20200228techkorea-coronavirus-tracking-appsindexhtml
United Kingdom
Hern Alex and Kate Proctor 2020 ldquoUK May Ditch NHS Contact-Tracing App for Apple and Google Modelrdquo The Guardian May 7 2020 httpswwwtheguardiancomtechnology2020may07uk-may-ditch-nhs-contact- tracing-app-for-apple-and-google-model
Lovejoy Ben 2020 ldquoHands-on with UKrsquos NHS Contact Tracing App as the Test Goes Liverdquo 9-5 Mac Blog May 7 2020 https9to5maccom2020 0507nhs-contact-tracing
ldquoCoronavirus Test Track and Trace Plan Launched on Isle of Wightrdquo Depart-ment of Health and Social Care Press Release May 4 2020 httpswww govukgovernmentnewscoronavirus-test-track-and-trace-plan-launched- on-isle-of-wight
Leprince-Ringuet Daphne 2020 ldquoContact-Tracing Apps Why the NHS Said No to Apple and Googlersquos Planrdquo ZDNet April 28 2020 httpswww zdnetcomarticlecontact-tracing-apps-why-the-nhs-said-no-to-apple- and-googles-plan
Specific Digital ProductsApps
COVID-19
Vota Wayan 2020 Additional Proposed Coronavirus Solutions Google Doc-ument httpsdocsgooglecomspreadsheetsd15hkhdtGNzx7oHkO8Y 2MOiY83JsHjqxL4MhMGvlA_J6Ieditgid=579623365
Apple and Google nd ldquoPrivacy-Preserving Contact Tracingrdquo Apple Accessed May 19 2020 httpswwwapplecomcovid19contacttracing
116 Resources
Woodhams Samuel 12 May 2020 ldquoCOVID-19 Digital Rights Trackerrdquo Top10VPN httpswwwtop10vpncomresearchinvestigationscovid-19-digital-rights-tracker
Starobinski David and Johannes Becker 30 April 2020 ldquoHow Apple and Goo-gle Will Let Your Phone Warn You If Yoursquove Been Exposed to the Corona-virusrdquo The Conversation Accessed May 21 2020 httptheconversation comhow-apple-and-google-will-let-your-phone-warn-you-if-youve-been-exposed-to-the-coronavirus-136597
Morrison Sara 2020 ldquoThe Apple-Google Contact Tracing Tool Gets a Beta Release and a New Risk Level Featurerdquo Vox April 24 2020 httpswwwvoxcomrecode202042421234420apple-google-contact-tracing- exposure-notification-update
Commonwealth Centre for Digital Health 9 April 2020 ldquo[Webinar] CWCDH Digital Response to COVID-19rdquo 5040 httpswwwyoutubecomwatchv=ZyE_KRWLtC8ampfeature=youtube
Johns Hopkins Medicine 23 April 2020 ldquoJohns Hopkins Medicine Remote Monitoring Program for Faculty Staff and Providers Exposed to COVID-19rdquo Johns Hopkins Medicine Occupational Health Services httpswww hopkinsmedicineorghsecovid19_emocha
MIT nd ldquoPrivate Kit Safe Paths Privacy-by-Design Covid19 Solutions Using GPS+Bluetooth for Citizens and Public Health Officialsrdquo Safepaths Accessed May 20 2020 httpssafepathsmitedu
Related Disease Detection Projects
ldquoCOVID Controlrdquo Google Play accessed May 21 2020 httpsplaygoogle comstoreappsdetailsid=jhueduJohnsHopkinsCOVIDControl
ldquoCOVID Symptom Trackerrdquo Created by Massachusetts General Hospital the Harvard TH Chan School of Public Health Kingrsquos College London and Stanford University School of Medicine accessed May 21 2020 httpscovidjoinzoecomus
ldquoSee How Your Community Is Moving around Differently Due to COVID-19rdquo Google COVID-19 Mobility Reports accessed May 21 2020 httpswww googlecomcovid19mobility
Drew David A Long H Nguyen Claire J Steves Cristina Menni Maxim Freydin Thomas Varsavsky Carole H Sudre M Jorge Cardoso Sebastien Ourselin Jonathan Wolf Tim D Spector Andrew T Chan and COPE Consortium 2020 ldquoRapid Implementation of Mobile Technology for Real-Time Epidemiology of COVID-19rdquo Science 05 May 2020 DOI 101126scienceabc0473
Resources 117
Tress Luke 2020 ldquoMaccabi Medial EarlySign Develop Algorithm to Identify High-Risk COVID-19 Casesrdquo Times of Israel April 22 2020 httpswwwtimesofisraelcommaccabi-medial-earlysign-develop-algo rithm-to-identify-high-risk-covid-19-cases
ldquoPCR Diagnostic Testing for SARS-CoV-2rdquo Center for Health Security Johns Hopkins Bloomberg School of Public Health last modified April 17 2020 httpswwwcenterforhealthsecurityorgresourcesCOVID-19COVID-19-fact-sheets200130-nCoV-diagnostics-factsheetpdf
Radin Jennifer M Nathan E Wineinger Eric J Topol and Steven R Steinhubl 2020 ldquoHarnessing Wearable Device Data to Improve State-Level Real-Time Surveillance of Influenza-Like Illness in the USA A Population-Based Studyrdquo The Lancet January 16 2020 DOIhttpsdoiorg101016S2589-7500(19)30222-5
Polling
Hargittai Eszter Minh Hao Nguyen Jaelle Fuchs Jonathan Gruber Will Marler Amanda Hunsaker and Goumlkccedile Karaoglu 2020 ldquoCovid-19 Study on Digital Media and the Coronavirus Pandemicrdquo Internet Use and Society Division Institute of Communication and Media Research University of Zurich httpwebuseorgcovid
Navigator Research 2020 ldquoPublic Opinion on Coronavirus Navigator Updaterdquo Navigating Coronavirus (blog) May 21 2020 https navigatorresearchorgnavigating-coronavirus
Russonello Giovanni 2020 ldquoBig Government For Now Most Americans Say Bring It Onrdquo The New York Times May 1 2020 sec US httpswww nytimescom20200501uspoliticscoronavirus-spending-pollshtml
ldquoWashington Post-University of Maryland National Pollrdquo 2020 Washington Post April 21 2020 httpswwwwashingtonpostcomcontext washington-post-university-of-maryland-national-poll-april-21-26-20203583b4e9-66be-4ed6-a457-f6630a550ddf
Kirzinger Ashley Liz Hamel Cailey Muntildeana Audrey Kearney and Mollyann Brodie 2020 ldquoKFF Health Tracking Poll ndash Late April 2020 Coronavirus Social Distancing and Contact Tracingrdquo Kaiser Family Foundation (blog) April 24 2020 httpswwwkfforgcoronavirus-covid-19issue-briefkff-health-tracking-poll-late-april-2020
NYC DOH 2020 ldquoAge-Adjusted Rates of Lab Confirmed COVID-19 Non-hospitalized Cases Estimated Non-Fatal Hospitalized Cases and Persons Known to Have Died per 100000 by RaceEthnicity Grouprdquo New York
118 Resources
City Department of Health httpswww1nycgovassetsdohdownloadspdfimmcovid-19-deaths-race-ethnicity-04242020-1pdf
Anderson Monica and Brooke Auxier 2020 ldquoMost Americans Donrsquot Think Cellphone Tracking Will Help Limit COVID-19 Are Divided on Whether Itrsquos Acceptablerdquo Pew Research Center (blog) April 16 2020 httpswwwpewresearchorgfact-tank20200416most-americans-dont-think-cellphone-tracking-will-help-limit-covid-19-are-divided-on-whether-its-acceptable
Daly Kyle 2020 ldquoExclusive Americans Wary of Giving up Data to Fight Coronavirusrdquo Axios April 3 2020 httpswwwaxioscomexclusive- americans-wary-of-giving-up-data-to-fight-coronavirus-330fc1d9-8b99-4a51-871b-25ee0e0591f2html
Elliott Douglas J Ana Kreacic Lorenzo Milans del Bosch and Lisa Quest nd ldquoData-Sharing in the Time of Coronavirusrdquo Oliver Wyman Forum Accessed May 19 2020 httpswwwoliverwymanforumcomfuture-of-data2020aprdata-sharing-in-the-time-of-coronavirushtml
Auxier Brooke Lee Rainie Monica Anderson Andrew Perrin Madhu Kumar and Erica Turner 2019 ldquoAmericans and Privacy Concerned Confused and Feeling Lack of Control Over Their Personal Informationrdquo Pew Research
Center Internet Science amp Tech (blog) November 15 2019 httpswww pewresearchorginternet20191115americans-and-privacy-concerned- confused-and-feeling-lack-of-control-over-their-personal-information
EPIC 2020 ldquoPublic Opinion on Privacyrdquo Electronic Privacy Information Center January 22 2020 httpsepicorgprivacysurvey
Pew Research Center 2020 ldquoDemographics of Mobile Device Ownership and Adoption in the United Statesrdquo Internet amp Technology Pew Research Cen-ter httpswwwpewresearchorginternetfact-sheetmobile
Rodrigues Rafaela Alina Husain Amanda Couture-Carron Leslye E Orloff and Nawal H Ammar 2018 ldquoPromoting Access to Justice for Immigrant and Limited English Proficient Crime Victims in an Age of Increased Immi-gration Enforcement Initial Report from a 2017 National Surveyrdquo Wash-ington DC National Immigrant Womenrsquos Advocacy Project American University Washington College of Law httpniwaplibrarywclamericaneduwp-contentuploadsImmigrant-Access-to-Justice-National-Reportpdf
Center for Survey Measurement (CSM) MEMORANDUM FOR Associate Directorate for Research and Methodology (ADRM) 2017 ldquoRespondent Confidentiality Concernsrdquo September 20 2017 httpswww2censusgovcacnacmeetings2017-11Memo-Regarding-Respondent-Confidentiality- Concernspdf
Resources 119
Pew Research Center 2017 ldquoUS Muslims Concerned About Their Place in Society but Continue to Believe in the American Dreamrdquo Religion amp Public Life Pew Research Center httpswwwpewforumorg20170726findings-from-pew-research-centers-2017-survey-of-us-muslims
Popular Press
Dixit Pranav 2020 ldquoIndiarsquos Contact Tracing App Is All But Mandatory So This Programmer Hacked It So That He Always Appears Saferdquo BuzzFeed News May 12 2020
Mills Rodrigo Chris 2020 ldquoDigital Contact Tracing Is Becoming Available but Is It Effectiverdquo The Hill May 7 2020 httpsthehillcompolicytechnology 496498-digital-contact-tracing-is-becoming-available-but-is-it-effective
Ingram David 2020 ldquoApple Google Push Makers of Coronavirus Apps Not to Record User Locationrdquo May 4 2020 NBC News Accessed May 19 2020 httpswwwnbcnewscomtechtech-newscoronavirus-apps-won-t-be-able- record-users-location-apple-n1199586
McMinn Sean 2020 ldquoMobile Phone Data Show More Americans Are Leav- ing Their Homes Despite Ordersrdquo NPR May 1 2020 httpswwwnpr org20200501849161820mobile-phone-data-show-more-americans- are-leaving-their-homes-despite-orders
Valentino-DeVries Jennifer Natasha Singer and Aaron Krolik 2020 ldquoA Scram-ble for Virus Apps That Do No Harmrdquo The New York Times April 29 2020 sec Business httpswwwnytimescom20200429 businesscoronavirus-cellphone-apps-contact-tracinghtml
Giglio Mike 2020 ldquoWould You Sacrifice Your Privacy to Get Out of Quaran-tinerdquo The Atlantic April 22 2020 httpswwwtheatlanticcompoliticsarchive202004coronavirus-pandemic-privacy-civil-liberties-911609172
Bradshaw Tim 2020 ldquo2 Billion Phones Cannot Use Google and Apple Con-tact-Tracing Tech | Ars Technicardquo Ars Technica April 20 2020 httpsarstechnicacomtech-policy2020042-billion-phones-cannot-use-google-and-apple-contract-tracing-tech
Reston Maeve Kristina Sgueglia and Cheri Mossburg 2020 ldquoGovernors on East and West Coasts Form Pacts to Decide When to Reopen Economiesrdquo CNN Politics April 13 2020 httpswwwcnncom20200413politicsstates-band-together-reopening-plansindexhtml
Cochrane Emily Claire Cain Miller and Jim Tankersley 2020 ldquoTrump Admin-istration Scales Back Paid Leave in Coronavirus Relief Lawrdquo The New York Times April 2 2020 sec US httpswwwnytimescom20200402uspoliticscoronavirus-paid-leavehtml
120 Resources
Kim Nemo 2020 ldquolsquoMore Scary than Coronavirusrsquo South Korearsquos Health Alerts Expose Private Livesrdquo The Guardian March 6 2020 sec World news httpswwwtheguardiancomworld2020mar06more-scary- than-coronavirus-south-koreas-health-alerts-expose-private-lives
Hickey Matt 2014 ldquoCarriers Can Now Install Apps On Android Handsets Without Customersrsquo Permissionrdquo Forbes December 1 2014 httpswww forbescomsitesmatthickey20141201carriers-can-now-install-apps-on- android-handsets-without-customers-permission
Commentaries
Giubilini Alberto 2020 ldquoContact-Tracing Apps and the Future COVID-19 Vaccination Should Be Compulsory Social Technological and Pharmaco-logical Immunisationrdquo Practical Ethics (blog) May 6 2020 httpblo gpracticalethicsoxacuk202005contact-tracing-apps-and-the-future- covid-19-vaccination-should-be-compulsory-social-technological-and- pharmacological-immunisation
Landau Susan Christy Lopez and Laura Moy 2020 ldquoThe Importance of Equity in Contact Tracingrdquo Lawfare (blog) May 1 2020 httpswww lawfareblogcomimportance-equity-contact-tracing
Schaefer G Owen and Angela Ballantyne 2020 ldquoDownloading COVID-19 Contact Tracing Apps Is a Moral Obligationrdquo Journal of Medical Ethics
Blog (blog) May 4 2020 httpsblogsbmjcommedical-ethics20200504downloading-covid-19-contact-tracing-apps-is-a-moral-obligation
OrsquoNeill Patrick Howell 2020 ldquoFive Things We Need to Do to Make Contact Tracing Really Workrdquo MIT Technology Review April 28 2020 httpswwwtechnologyreviewcom202004281000714five-things-to-make- contact-tracing-work-covid-pandemic-apple-google
Doffman Zak 2020 ldquoCOVID-19 Contact Tracing Why Apple And Google Canrsquot Make This Workrdquo Forbes April 27 2020 httpswwwforbescomsiteszakdoffman20200427this-is-the-contact-tracing-worry-even-apple- and-google-cant-resolve
All Tech is Human 2020 The Ethics of Contact Tracing for COVID-19 httpswwwyoutubecomwatchv=59mKUAVDhdkampt=626s
Canca Cansu 2020 ldquoWhy lsquoMandatory Privacy-Preserving Digital Contact Tracingrsquo Is the Ethical Measure against COVID-19rdquo Medium May 4 2020 httpsmediumcomcansucancawhy-mandatory-privacy-preserv-ing-digital-contact-tracing-is-the-ethical-measure-against-covid-19-a0d-143b7c3b6
Goodman Bryce 2020 ldquoCOVID and Contact Tracing When Social Justice
Resources 121
Demands Mass Surveillancerdquo Medium April 10 2020 httpsmediumcombwgoodmancovid-and-contact-tracing-when-social-justice-demands-mass-surveillance-18d419b8cc5
Gray Rosie and Caroline Haskins 2020 ldquoThey Were Opposed To Govern-ment Surveillance Then The Coronavirus Pandemic Beganrdquo BuzzFeed News March 30 2020 httpswwwbuzzfeednewscomarticlerosiegraythey-were-opposed-to-government-surveillance-then-the
Cegłowski Maciej 2020 ldquoWe Need A Massive Surveillance Programrdquo Idle
Words (blog) March 23 2020 httpsidlewordscom202003we_need_ a_massive_surveillance_programhtm
Academic Literature
COVID-19 Specific
Abeler J Baumlcker M Buermeyer U Zillessen H (2020) COVID-19 Contact Tracing and Data Protection Can Go Together JMIR mHealth and uHealth 8(4) e19359 doi 10219619359
Altmann S Milsom L Zillessen H et al (2020) Acceptability of App-Based Contact Tracing for COVID-19 Cross-Country Survey Evidence Preprint
Berke A Bakker M Vepakomma P Larson K Pentland A (2020) Assessing Disease Exposure Risk with Location Data A Proposal for Cryptographic Preservation of Privacy arXiv arXiv200314412ndashMarch 2020
Bonsail D Parker M Fraser C (2020) Sustainable Containment of COVID-19 Using Smartphones in China Scientific and Ethical Underpinnings for Implementation of Similar Approaches in Other Settings Unpublished working paper
Bradshaw WJ Alley EC Huggins JH Lloyd AL Esvelt KM (2020) Bidirectional Contact Tracing Is Required for Reliable COVID-19 Control Preprint via MedRxiv
Braithwaite I Callender T Bullock M Aldridge R (2020) Automated and Semi-Automated Contact Tracing Protocol for a Rapid Review of Avail-able Evidence and Current Challenges to Inform the Control of COVID-19 Preprint via medRxiv doi httpsdoiorg1011012020041420063636
Bulchandani Bannerjee V Shivam S Moudgalya S Sondhi SL (2020) Digital Herd Immunity and COVID-19 Preprint via medRxiv doi httpsdoiorg1011012020041520066720
Cheng H Jian S Liu D (2020) Contact Tracing Assessment of COVID-19 Transmission Dynamics in Taiwan and Risk at Different Exposure Periods Before and After Symptom Onset JAMA Internal Medicine doi101001jamainternmed20202020
122 Resources
Cho H Ippolito D Yu YW (2020) Contact Tracing Mobile Apps for COVID-19 Privacy Considerations and Related Trade-offs httpsarxivorgpdf200311511pdf
Devakumar D Geordan S Bhopal SS Abubakar I (2020) Racism and dis-crimination in COVID-19 responses The Lancet 395(10231) 1194 doi 101016S0140-6736(20)30792-3
Drew D Nguyen L Steves C et al (2020) Rapid Implementation of Mobile Technology for Real-Time Epidemiology of COVID-19 Science published online May 5 2020
Ferretti L Wymant C Kendall M et al (2020) Quantifying SARS-CoV-2 Trans-mission Suggests Epidemic Control with Digital Contact Tracing Science doi 101126scienceabb6936
Fraser C Abeler-Doumlrner L Ferretti L et al (2020) Digital Contact Tracing Comparing the Capabilities of Centralised and Decentralised Data Archi-tectures to Effectively Suppress the COVID-19 Epidemic While Maximizing Freedom of Movement and Maintaining Privacy Preprint
Leith DJ Farrell S (2020) Coronavirus Contact Tracing Evaluating The Poten-tial Of Using Bluetooth Received Signal Strength For Proximity Detection Preprint
Jayant Limaye R Sauer M Ali J et al (2020) Building Trust While Influencing Online COVID-19 Content iIn the Social Media World The Lancet Digital Health
Mahmood S Hasan K Colder Carras M Labrique A (2020) Global Prepared-ness Aagainst COVID-19 We Must Leverage the Power of Digital Health JMIR Public Health Surveill 20206(2)e18980 DOI 10219618980
Mello M Wang CJ (2020) Ethics and Governance for Digital Disease Surveil-lance Science 11 May 2020 eabb9045 DOI 101126scienceabb9045
Park S Jeehyun Choi G Ko H (2020) Information TechnologyndashBased Tracing Strategy in Response to COVID-19 in South KoreamdashPrivacy Controversies JAMA doi101001jama20206602
Parker M Fraser C Abeler-Doumlrner L Bonsall D (2020) Ethics of Instantaneous Contract Tracing Using Mobile Phone Apps in the Control of the COVID-19 Pandemic Journal of Medical Ethics Published Online May 4 2020
Ethics and Digital Disease Detection
Aiello A Renson A Civich P (2020) Social Mediandash and Internet-Based Dis-ease Surveillance for Public Health Annual Review of Public Health 41 101ndash118 doi 101146annurev-publhealth-040119-094402
Ali J DiStefano M Coates McCall I et al (2019) Ethics of Mobile Phone Sur-
Resources 123
veys to Monitor Non-Communicable Disease Risk Factors in Low- and Middle-Income Countries A Global Stakeholder Survey Global Public Health 14(8) 1167ndash1181
Ali J Labrique A Gionfriddo K et al (2017) Ethics Considerations in Global Mobile Phone-Based Surveys of Noncommunicable Diseases A Concep-tual Exploration Journal of Medical Internet Research 19(5) e110 doi 102196jmir7326
Brownstein J Freifeld C Madoff L (2009) Digital Disease DetectionndashHar-nessing the Web for Public Health Surveillance New England Journal of Medicine 360(21) 2153ndash2157 doi 101056NEJMp0900702
Danquah LO Hasham N MacFarlane M et al (2019) Use of a Mobile Application for Ebola Contact Tracing and Monitoring in Northern Sierra Leone A Proof-of-Concept Study BMC Infectious Diseases 19 810
Degeling C Carter S van Oijen A et al (2020) Community Perspectives on the Benefits and Risks of Technologically Enhanced Communicable Disease Surveillance Systems A Report on Four Community Juries BMC Medical Ethics 21 31 doi 101186s12910-020-00474-6
DeJong B Badou G Luten J et al (2019) Ethical Considerations for Movement Mapping to Identify Disease Transmission Hotspots Emerging Infectious Diseases 25(7) e181421 doi 103201eid2507181421
Denecke K (2017) An Ethical Assessment Model for Digital Disease Detec-tion Technologies Life Sciences Society and Policy 13 16 doi 101186s40504-017-0062-x
Genevieve LD Martani A Wangmo T et al (2019) Participatory Disease Surveillance Systems Ethical Framework Journal of Medical Internet Research 21(5) e12273 doi10219612273
Gilbert G Degeling C Johnson J (2017) Communicable Disease Surveillance Ethics in the Age of Big Data and New Technology Asian Bioethics Review 11 173-187 doi 101007s41649-019-00087-1
Iwaya LH Li J Fischer-Hubner S et al (2019) E-Consent for Data Privacy Consent Management for Mobile Health Technologies in Public Health Surveys and Disease Surveillance Studies in health technology and infor-matics 264 1223-1227 doi 103233SHTI190421
Kostkova P (2018) Disease Surveillance Data Sharing for Public Health The Next Ethical Frontiers Life Sciences Society and Policy 14 16 doi 101186s40504-018-0078-x
Mahmood S Hasan K Colder Carras M Labrique A (2020) Global Prepared-ness against COVID-19 We Must Leverage the Power of Digital Health JMIR Public Health Surveillance 6(2)e18980
124 Resources
Smolinski MS Crawley AW Baltrusaitis K et al (2015) Flu Near You Crowd-sourced Symptom Reporting Spanning 2 Influenza Seasons AJPH 105(10) 2124ndash2130
Wojcik O Brownstein J Chunara R Johansson M (2014) Public Health for the People Participatory Infectious Disease Surveillance in the Digital Age Emerging Themes in Epidemiology 11 7 doi 1011861742-7622-11-7
Other Works
Bernstein J Holroyd TA Atwell JE et al (2019) Rockland Countyrsquos Proposed Ban against Unvaccinated Minors Balancing Disease Control Trust and Liberty Vaccine 37(30) 3933ndash3935
Berry SM Petzold EA Dull P et al (2017) A Response Adaptive Randomiza-tion Platform Trial for Efficient Evaluation of Ebola Virus Treatments A Model for Pandemic Response Clinical Trials 13(1) 22ndash30 doi 101177 1740774515621721
Beukenhorst AL Schultz DM McBeth J (2017) Using Smartphones for Research outside Clinical Settings How Operating Systems App Develop-ers and Users Determine Geolocation Data Quality in mHealth Studies MEDINFO 2017 Precision Healthcare through Informatics
Bourne P (2015) Confronting the Ethical Challenges of Big Data in Public Health Plos Computational Biology 11(2) e1004073 doi 101371journal pcbi1004073
Doerr M Suver C Wilbanks J (2016) Developing a Transparent Par ticipant-Navigated Electronic Informed Consent for Mobile-Mediated Research (April 22 2016) Available at SSRN httpsssrncomabstract=2769129 or httpdxdoiorg102139ssrn2769129
Dredze M Paul MJ Bergsma S Tran H (2013) Carmen A Twitter Geolocation System with Applications to Public Health Expanding the Boundaries of Health Informatics Using Artificial Intelligence Papers from the AAAI 2013 Workshop
Eckhoff PA Tatem AJ (2015) Digital methods in epidemiology can transform disease controlInternational Health Volume 7 Issue 2 March 2015 Pages 77ndash78 httpsdoiorg101093inthealthihv013
Edelstein M Lee L Herten-Crabb A Heymann D Harper D (2018) Strength-ening Global Public Health Surveillance through Data and Benefit Sharing Emerging Infectious Diseases 24(7) 1324ndash1330 doi 103201eid2407151830
Faden R Beauchamp T (1986) A History and Theory of Informed Consent Oxford University Press ISBN 9780199748655
Resources 125
Fairchild A Bayer R (2004) Ethics and the Conduct of Public Health Surveil-lance Science 303(5658) 631ndash632
Flanagan M Howe DC Nissenbaum H (2008) Embodying Values in Technol-ogy Theory and Practice In Information Technology and Moral Philoso-phy van den Hoven J amp Weckert J (eds) Cambridge Cambridge Univer-sity Press 322ndash353
Fraccaro P Beukenhorst A Sperrin M et al (2019) Digital Biomarkers from Geolocation Data in Bipolar Disorder and Schizophrenia A Systematic Review Journal of the American Medical Informatics Association Volume 26 Issue 11 November 2019 Pages 1412ndash1420
Furlanello C Merler S Menegon S et al (2002) New WEBGIS Technologies for Geo-location of Epidemiological Data An Application for the Sur-veillance of the Risk of Lyme borreliosis Disease In Giornale Italiano di Aritmologia e Cardiostimolazione Special Issue Proceedings of the 15th Int Congress ldquoThe New Frontiers of Arrhythmiasrdquo v 5 n1 Mar 2002 241ndash245
Gibson DG Wosu AC Pariyo GW et al (2019) Effect of Airtime Incentives on Response and Cooperation Rates in Non-communicable Disease Interactive Voice Response Surveys Randomised Controlled Trials in Bangladesh and Uganda BMJ Global Health 4(5)
Knobel C Bowker GC (2011) Computing Ethics Values in Design Communi-cations of the acm 54(7) 26ndash28
Lee R Cui RR Muessig KE et al (2015) Incentivizing HIVSTI Testing A Sys-tematic Review of the Literature AIDS and Behavior 18(5) 905ndash912 doi 101007s10461-013-0588-8
Lorchan LT Wyatt J (2014) mHealth and Mobile Medical Apps A Frame-work to Assess Risk and Promote Safer Use Journal of Medical Internet Research 16(9) e210 doi 102196jmir3133
Mathews S McShea M Hanley C et al (2019) Digital Health A Path to Vali-dation Digital Medicine 238 doi 101038s41746-019-0111-3
Moore S Tasse A Thorogood A et al (2017) Consent Processes for Mobile App Mediated Research Systematic Review JMIR mHealth amp uHealth 5(8) e126 DOI 102196mhealth7014
Pallman P Bedding AW Choodari-Oskooei B et al (2018) Adaptive Designs in Clinical Trials Why Use Them and How to Run and Report Them BMC Medicine 16(29)
Rennie S Buchbinder M Juengst E et al (2020) Scraping the Web for Public Health Gains Ethical Considerations from a lsquoBig Datarsquo Research Project on HIV and Incarceration Public Health Ethics doi 101093phephaa006
126 Resources
Rithalia A McDaid C Suekarran S (2009) Impact of Presumed Consent for Organ Donation on Donation Rates A Systematic Review BMJ 2009 338
Singer E Ye C (2012) The Use and Effects of Incentives in Surveys The ANNALS of the American Academy of Political and Social Science 645(1) 112ndash141
Vayena E Blassime A (2018) Health Research with Big Data Time for Systemic Oversight J Law Med Ethics 2018 Mar 46(1) 119ndash129
Vayena E Mastroianni A Kahn J (2012) Ethical Issues in Health Research with Novel Online Sources American Journal of Public Health 102(12) 2225ndash2230 doi 102105AJPH2012300813
Vayena E Salatheacute M Madoff L Brownstein J (2015) Ethical Challenges of Big Data in Public Health Plos Computational Biology 11(2) e1003904 doi 101371journalpcbi1003904
Xafis V Schaefer GO Labude MK et al (2019) An Ethics Framework for Big Data in Health and Research Asian Bioethics Review volume 11 pages 227ndash254(2019)
Works Cited
127
Africa CDC 2020 ldquoGuidance on Contact Tracing for COVID-19 Pandemicrdquo Manuals Guidelines amp Frameworks African Union Africa CDC httpsafricacdcorgdownloadguidance-on-contact-tracing-for-covid-19- pandemic
Aiello Allison E Audrey Renson and Paul N Zivich 2020 ldquoSocial Mediandash and Internet-Based Disease Surveillance for Public Healthrdquo Annual Review of Public Health 41 (1) 101ndash18 httpsdoiorg101146annurev-publhealth-040119-094402
Ali Joseph Michael J DiStefano Iris Coates McCall Dustin G Gibson Gulam Muhammed Al Kibria George W Pariyo Alain B Labrique and Adnan A Hyder 2019 ldquoEthics of Mobile Phone Surveys to Monitor Non-Communicable Disease Risk Factors in Low- and Middle-Income Countries A Global Stakeholder Surveyrdquo Global Public Health 14 (8) 1167ndash81 httpsdoiorg1010801744169220191566482
Ali Joseph Alain B Labrique Kara Gionfriddo George Pariyo Dustin G Gibson Bridget Pratt Molly Deutsch-Feldman and Adnan A Hyder 2017 ldquoEthics Considerations in Global Mobile Phone-Based Surveys of Noncommunicable Diseases A Conceptual Explorationrdquo Journal of Medi-cal Internet Research 19 (5) e110 httpsdoiorg102196jmir7326
Allen Danielle Julius Krein Ganesh Sitaraman and E Glen Weyl 2020 ldquoNational Covid-19 Testing Action Planrdquo The Rockefeller Foundation httpswwwrockefellerfoundationorgwp-contentuploads202004TheRockefellerFoundation_WhitePaper_Covid19_4_22_2020pdf
Altmann Samuel Luke Milsom Hannah Zillessen Raffaele Blasone Frederic Gerdon Ruben Bach Frauke Kreuter Daniele Nosenzo Severine Tous-saert and Johannes Abeler 2020 ldquoAcceptability of App-Based Contact Tracing for COVID-19 Cross-Country Survey Evidencerdquo MedRxiv May 2020050520091587 httpsdoiorg1011012020050520091587
Anderson Monica and Brooke Auxier 2020 ldquoMost Americans Donrsquot Think Cellphone Tracking Will Help Limit COVID-19 Are Divided on Whether
128 Works Cited
Itrsquos Acceptablerdquo Pew Research Center (blog) April 16 2020 https wwwpewresearchorgfact-tank20200416most-americans-dont-think-cellphone-tracking-will-help-limit-covid-19-are-divided-on-whether-its- acceptable
Anderson Monica and Andrew Perrin 2017 ldquoDisabled Americans Less Likely to Use Technologyrdquo Pew Research Center (blog) April 7 2017 httpswwwpewresearchorgfact-tank20170407disabled-americans- are-less-likely-to-use-technology
Apple and Google nd ldquoPrivacy-Preserving Contact Tracingrdquo Apple Accessed May 19 2020 httpswwwapplecomcovid19contacttracing
ASTHO 2012 ldquoPublic Health Collection Use Sharing and Protection of Informationrdquo Issue Brief ASTHO Legal Preparedness Series Association of State and Territorial Health Officials httpswwwasthoorgProgramsPreparednessPublic-Health-Emergency-LawPublic-Health-and- Information-Sharing-ToolkitCollection-Use-Sharing-and-Protection- Issue-Brief
Auxier Brooke Lee Rainie Monica Anderson Andrew Perrin Madhu Kumar and Erica Turner 2019 ldquoAmericans and Privacy Concerned Confused and Feeling Lack of Control Over Their Personal Informationrdquo Pew Research Center Internet Science amp Tech (blog) November 15 2019 httpswwwpewresearchorginternet20191115americans-and-privacy- concerned-confused-and-feeling-lack-of-control-over-their-personal-infor mation
Barth Susanne and Menno D T de Jong 2017 ldquoThe Privacy Paradox ndash Inves-tigating Discrepancies between Expressed Privacy Concerns and Actual Online Behavior ndash A Systematic Literature Reviewrdquo Telematics and Infor-matics 34 (7) 1038ndash58 httpsdoiorg101016jtele201704013
Bernstein Justin Taylor A Holroyd Jessica E Atwell Joseph Ali and Rupali J Limaye 2019 ldquoRockland Countyrsquos Proposed Ban against Unvaccinated Minors Balancing Disease Control Trust and Libertyrdquo Vaccine June httpsdoiorg101016jvaccine201905093
Berry Scott M Elizabeth A Petzold Peter Dull Nathan M Thielman Coleen K Cunningham G Ralph Corey Micah T McClain et al 2016 ldquoA Response Adaptive Randomization Platform Trial for Efficient Evalua-tion of Ebola Virus Treatments A Model for Pandemic Responserdquo Clinical Trials (London England) 13 (1) 22ndash30 httpsdoiorg101177 1740774515621721
Beukenhorst AL DM Schultz J McBeth R Lakshminarayana JC Sergeant and WG Dixon 2017 ldquoUsing Smartphones for Research Outside Clinical
Works Cited 129
Settings How Operating Systems App Evelopers and Users Determine Geolocation Data Quality in MHealth Studiesrdquo In MEDINFO 2017 Precision Healthcare through Infomatics 10ndash14 IOS Press httpebooks iospressnlvolumearticle48095
Bradshaw Tim 2020 ldquo2 Billion Phones Cannot Use Google and Apple Con-tact-Tracing Tech | Ars Technicardquo Ars Technica April 20 2020 httpsarstechnicacomtech-policy2020042-billion-phones-cannot-use-google-and-apple-contract-tracing-tech
Canca Cansu 2020 ldquoWhy lsquoMandatory Privacy-Preserving Digital Contact Tracingrsquo Is the Ethical Measure against COVID-19rdquo Medium May 4 2020 httpsmediumcomcansucancawhy-mandatory-privacy- preserving-digital-contact-tracing-is-the-ethical-measure-against- covid-19-a0d143b7c3b6
Cavalier Robert J ed 2011 Approaching Deliberative Democracy Theory and Practice Carnegie Mellon University Press
Cavoukian Ann 2010 ldquoPrivacy by Design The 7 Foundational Principlesrdquo Ontario Canada Information and Privacy Commissioner of Canada httpsiapporgmediapdfresource_centerpbd_implement_7found_ principlespdf
CDC 2005 ldquoVI Plan for Surveillance of Contacts of SARS Case Supplement B SARS Surveillancerdquo Centers for Disease Control and Prevention httpswwwcdcgovsarsguidanceb-surveillancecontactshtml
mdashmdashmdash 2020a ldquoContact Tracingrdquo Get and Keep America Open Supporting States Tribes Localities and Territories Centers for Disease Control and Prevention httpswwwcdcgovcoronavirus2019-ncovphpopen-americacontact-tracinghtml
mdashmdashmdash 2020b ldquoContact Tracing Part of a Multipronged Approach to Fight the COVID-19 Pandemicrdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention httpswwwcdcgovcoronavirus2019- ncovphpprinciples-contact-tracinghtml
mdashmdashmdash 2020c ldquoInterim Guidance for Businesses and Employersrdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention httpswwwcdcgovcoronavirus2019-ncovcommunityguidance-business- responsehtml
mdashmdashmdash 2020d ldquoOperational Considerations for the Identification of Healthcare Workers and Inpatients with Suspected COVID-19 in Non-US Healthcare Settingsrdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention httpswwwcdcgovcoronavirus2019-ncovhcpnon-us-settingsguidance-identify-hcw-patientshtml
130 Works Cited
mdashmdashmdash 2020e ldquoPreliminary Criteria for the Evaluation of Digital Contact Trac-ing Tools for COVID-19rdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention httpswwwcdcgovcoronavirus 2019-ncovdownloadsphpprelim-eval-criteria-digital-contact-tracingpdf
mdashmdashmdash 2020f ldquoLegal Authorities for Isolation and Quarantinerdquo Centers for Disease Control and Prevention httpswwwcdcgovquarantine aboutlawsregulationsquarantineisolationhtml
mdashmdashmdash 2020g ldquoDiscontinuation of Isolation for Persons with COVID-19 Not in Healthcare Settingsrdquo Interim Guidance Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention httpswww cdcgovcoronavirus2019-ncovhcpdisposition-in-home-patientshtml
mdashmdashmdash 2020h ldquoCOVID-19 Provisional CountsndashWeekly Updates by Select Demographic and Geographic Characteristicsrdquo CDC National Vital Statis-tics System httpswwwcdcgovnchsnvssvsrrcovid_weeklyindexhtm
Center for Survey Measurement (CSM) MEMORANDUM FOR Associate Directorate for Research and Methodology (ADRM) 2017 ldquoRespondent Confidentiality Concernsrdquo September 20 2017 httpswww2censusgovcacnacmeetings2017-11Memo-Regarding-Respondent-Confidentiality- Concernspdf
Cheng Hao-Yuan Shu-Wan Jian Ding-Ping Liu Ta-Chou Ng Wan-Ting Huang and Hsien-Ho Lin 2020 ldquoContact Tracing Assessment of COVID-19 Transmission Dynamics in Taiwan and Risk at Different Exposure Periods Before and After Symptom Onsetrdquo JAMA Internal Medicine May httpsdoiorg101001jamainternmed20202020
Cochrane Emily Claire Cain Miller and Jim Tankersley 2020 ldquoTrump Admin-istration Scales Back Paid Leave in Coronavirus Relief Lawrdquo The New York Times April 2 2020 sec US httpswwwnytimescom20200402uspoliticscoronavirus-paid-leavehtml
Crocker Andrew Kurt Opsahl and Bennett Cyphers 2020 ldquoThe Challenge of Proximity Apps For COVID-19 Contact Tracingrdquo Electronic Fron-tier Foundation April 10 2020 httpswwwefforgdeeplinks202004challenge-proximity-apps-covid-19-contact-tracing
Danquah Lisa O Nadia Hasham Matthew MacFarlane Fatu E Conteh Fatoma Momoh Andrew A Tedesco Amara Jambai David A Ross and Helen A Weiss 2019 ldquoUse of a Mobile Application for Ebola Contact Tracing and Monitoring in Northern Sierra Leone A Proof-of-Concept Studyrdquo BMC Infectious Diseases 19 (1) 810 httpsdoiorg101186s12879-019-4354-z
de Jong Bouke C Badou M Gaye Jeroen Luyten Bart van Buitenen Emman-
Works Cited 131
uel Andreacute Conor J Meehan Cian OrsquoSiochain et al 2019 ldquoEthical Considerations for Movement Mapping to Identify Disease Transmission Hotspotsrdquo Emerging Infectious Diseases 25 (7) httpsdoiorg103201eid2507181421
Devakumar Delan Geordan Shannon Sunil S Bhopal and Ibrahim Abu-bakar 2020 ldquoRacism and Discrimination in COVID-19 Responsesrdquo Lancet (London England) 395 (10231) 1194 httpsdoiorg101016S0140-6736(20)30792-3
Dixit Pranav 2020 ldquoIndiarsquos Contact Tracing App Is All But Mandatory So This Programmer Hacked It So That He Always Appears Saferdquo BuzzFeed News May 12 2020 httpswwwbuzzfeednewscomarticlepranavdixitindia-aarogya-setu-hacked
Doerr Megan Christine Suver and John Wilbanks 2016 ldquoDeveloping a Trans-parent Participant-Navigated Electronic Informed Consent for Mobile- Mediated Researchrdquo SSRN Scholarly Paper ID 2769129 Rochester NY Social Science Research Network httpsdoiorg102139ssrn2769129
Dredze Mark Michael J Paul Shane Bergsma and Hieu Tran 2013 ldquoCarmen A Twitter Geolocation System with Applications to Public Healthrdquo AAAI Workshops Workshops at the Twenty-Seventh AAAI Conference on Artifi-cial Intelligence httpswwwaaaiorgocsindexphpWSAAAIW13paperview70856497
Eckhoff Philip A and Andrew J Tatem 2015 ldquoDigital Methods in Epidemi-ology Can Transform Disease Controlrdquo International Health 7 (2) 77ndash78 httpsdoiorg101093inthealthihv013
EEOC 2020 ldquoWhat You Should Know About COVID-19 and the ADA the Rehabilitation Act and Other EEO Lawsrdquo US Equal Employment Opportunity Commission May 7 2020 httpswwweeocgovwyskwhat-you-should-know-about-covid-19-and-ada-rehabilitation-act-and-other-eeo-laws
Electronic Privacy Information Center Testimony to Congress 2020 ldquoEPIC to Congress Establish Privacy Safeguards for Digital Contact Tracingrdquo April 15 2020 httpsepicorgtestimonycongressEPIC-HEC-Contact-Tracing-Apr2020pdf
Elliott Douglas J Ana Kreacic Lorenzo Milans del Bosch and Lisa Quest 2020 ldquoData-Sharing in the Time of Coronavirusrdquo Oliver Wyman Forum Accessed May 19 2020 httpswwwoliverwymanforumcomfuture-of-data2020aprdata-sharing-in-the-time-of-coronavirushtml
Faden Ruth R and Tom L Beauchamp 1986 A History and Theory of Informed Consent Oxford University Press
132 Works Cited
FCC and FTC 2017 ldquoFCC-FTC Consumer Protection Memorandum of Understandingrdquo Federal Communications Commission and Federal Trade Commission httpswwwftcgovsystemfilesdocumentscooperation_agreements151116ftcfcc-moupdf
Ferretti Luca Chris Wymant Michelle Kendall Lele Zhao Anel Nurtay Lucie Abeler-Doumlrner Michael Parker David Bonsall and Christophe Fraser 2020 ldquoQuantifying SARS-CoV-2 Transmission Suggests Epidemic Control with Digital Contact Tracingrdquo Science 368 (6491) httpsdoiorg101126scienceabb6936
Fishkin James S and Peter Laslett eds 2003 Debating Deliberative Democ-racy Blackwell Publisher Ltd httpsonlinelibrarywileycomdoibook 1010029780470690734
Flanagan Mary Daniel C Howe and Helen Nissenbaum 2008 ldquoEmbodying Values in Technology Theory and Practicerdquo In Information Technology and Moral Philosophy edited by Jeroen van den Hoven and John Weckert 322ndash53 Cambridge University Press
Fraccaro Paolo Anna Beukenhorst Matthew Sperrin Simon Harper Jasper Palmier-Claus Shocircn Lewis Sabine N Van der Veer and Niels Peek 2019 ldquoDigital Biomarkers from Geolocation Data in Bipolar Disorder and Schizophrenia A Systematic Reviewrdquo Journal of the American Medical Informatics Association 26 (11) 1412ndash20 httpsdoiorg101093jamiaocz043
Fraser Christophe Lucie Abeler-Doumlrner Luca Ferretti Michael Parker Michelle Kendall and David Bonsall 2020 ldquoDigital Contact Tracing Comparing the Capabilities of Centralised and Decentralised Data Architectures to Effectively Suppress the COVID-19 Epidemic Whilst Maximising Freedom of Movement and Maintaining Privacyrdquo httpsgithubcomBDI-pathogenscovid-19_instant_tracingblobmasterCentralised20and20decen-tralised20systems20for20contact20tracingpdf
FTC 2012 ldquoProtecting Consumer Privacy in an Era of Rapid Changerdquo FTC Report Recommendations for Businesses and Policymakers Federal Trade Commission httpswwwftcgovsitesdefaultfilesdocumentsreports federal-trade-commission-report-protecting-consumer-privacy- era-rapid-change-recommendations120326privacyreportpdf
mdashmdashmdash 2020 ldquoPrivacy amp Data Security Update 2019rdquo Federal Trade Commis-sion httpswwwftcgovsystemfilesdocumentsreportsprivacy-data- security-update-20192019-privacy-data-security-report-508pdf
Furlanello Cesare Stefano Merler Stefano Menegon Sebastiano Mancuso and Gianni Bertiato 2002 ldquoNew WEBGIS Technologies for Geo-Location of
Works Cited 133
Epidemiological Data An Application for the Surveillance of the Risk of Lyme Borreliosis Diseaserdquo Giornale Italiano Di Aritmologia e Cardiosti-molazione Special Issue Proceedings of the 15th Int Congress ldquoThe New Frontiers of Arrhythmiasrdquo 5 (1) 241ndash45
Gan Nectar and David Culver 2020 ldquoChina Is Fighting the Coronavirus with a Digital QR Code Herersquos How It Worksrdquo CNN Business April 16 2020 httpswwwcnncom20200415asiachina-coronavirus-qr-code-intl-hnkindexhtml
Gibson Dustin G Adaeze C Wosu George William Pariyo Saifuddin Ahmed Joseph Ali Alain B Labrique Iqbal Ansary Khan Elizeus Rutebemberwa Meerjady Sabrina Flora and Adnan A Hyder 2019 ldquoEffect of Airtime Incentives on Response and Cooperation Rates in Non-Communicable Disease Interactive Voice Response Surveys Randomised Controlled Trials in Bangladesh and Ugandardquo BMJ Global Health 4 (5) e001604 httpsdoiorg101136bmjgh-2019-001604
Giglio Mike 2020 ldquoWould You Sacrifice Your Privacy to Get Out of Quarantinerdquo The Atlantic April 22 2020 httpswwwtheatlanticcompoliticsarchive202004coronavirus-pandemic-privacy-civil-liberties-911 609172
Guariglia Matthew 2020 ldquoThe Dangers of COVID-19 Surveillance Proposals to the Future of Protestrdquo Electronic Frontier Foundation April 29 2020 httpswwwefforgdeeplinks202004some-covid-19-surveillance-propos als-could-harm-free-speech-after-covid-19
Hadavas Chloe 2020 ldquoHow Effective Are Contact Tracing Appsrdquo Slate Maga-zine May 13 2020 httpsslatecomtechnology202005contact- tracing-apps-less-effective-icelandhtml
Hamilton Isobel Asher 2020 ldquoPoland Made an App That Forces Coronavirus Patients to Take Regular Selfies to Prove Theyrsquore Indoors or Face a Police Visitrdquo Business Insider March 23 2020 httpswwwbusinessinsidercompoland-app-coronavirus-patients-mandaotory-selfie-2020-3
Hargittai Eszter Minh Hao Nguyen Jaelle Fuchs Jonathan Gruber Will Marler Amanda Hunsaker and Goumlkccedile Karaoglu 2020 ldquoCovid-19 Study on Digital Media and the Coronavirus Pandemicrdquo Internet Use and Society Division Institute of Communication and Media Research University of Zurich httpwebuseorgcovid
Hargittai Eszter and Elissa Redmiles 2020 ldquoWill Americans Be Willing to Install COVID-19 Tracking Appsrdquo Scientific American Blog Net-work April 28 2020 httpsblogsscientificamericancomobservationswill-americans-be-willing-to-install-covid-19-tracking-apps
134 Works Cited
Hart Vi Divya Siddarth Bethan Cantrell Lila Tretikov Peter Eckersley John Langford Scott Leibrand et al 2020 ldquoOutpacing the Virus Dig-ital Response to Containing the Spread of COVID-19 While Mitigat-ing Privacy Risksrdquo Whitepaper 5 COVID-19 Rapid Response Impact Initiative Edmond J Safra Center for Ethics httpsdrivegooglecomfiled1vIN2AX-DDNW-S0aHq8xs0RJ2jkR_CckXview
Hemming K T P Haines P J Chilton A J Girling and R J Lilford 2015 ldquoThe Stepped Wedge Cluster Randomised Trial Rationale Design Anal-ysis and Reportingrdquo BMJ 350 (February) httpsdoiorg101136bmjh391
Hendrix Steve and Ruth Eglash 2020 ldquoIsrael Is Using Cellphone Surveil-lance to Warn Citizens You May Already Be Infectedrdquo Washington Post Accessed May 19 2020 httpswwwwashingtonpostcomworldmiddle_eastisrael-is-using-cellphone-surveillance-to-warn-citizens-you-may- already-be-infected2020031968267294-69e7-11ea-b199-3a9799c54 512_storyhtml
Heneghan Carl Jon Brassey and Tom Jefferson 2020 ldquoCOVID-19 What Proportion Are Asymptomaticrdquo Centre for Evidence-Based Medi-cine httpswwwcebmnetcovid-19covid-19-what-proportion-are- asymptomatic
HHS 2019 ldquoPublic Health Emergency Declarationrdquo Public Health Emergency Accessed May 19 2020 httpswwwphegovPreparednesslegalPagesphedeclarationaspx
Hickey Matt 2014 ldquoCarriers Can Now Install Apps On Android Handsets Without Customersrsquo Permissionrdquo Forbes December 1 2014 httpswww forbescomsitesmatthickey20141201carriers-can-now-install- apps-on-android-handsets-without-customers-permission
Hinch Robert Will Probert Anel Nurtay Michelle Kendall Chris Wymant Matthew Hall Katrina Lythgoe et al 2020 ldquoEffective Configurations of a Digital Contact Tracing App A Report to NHSXrdquo fileUsersameliahoodDownloadsReport20-20Effective20App20Configurations20(1)pdf
Ingram David 2020 ldquoApple Google Push Makers of Coronavirus Apps Not to Record User Locationrdquo May 4 2020 NBC News Accessed May 19 2020 httpswwwnbcnewscomtechtech-newscoronavirus-apps-won-t-be-able- record-users-location-apple-n1199586
Iwaya Leonardo H Jane Li Simone Fischer-Huumlbner Rose-Mharie Aringhlfeldt and Leonardo A Martucci 2019 ldquoE-Consent for Data Privacy Consent Management for Mobile Health Technologies in Public Health Surveys and
Works Cited 135
Disease Surveillancerdquo Studies in Health Technology and Informatics 264 (August) 1223ndash27 httpsdoiorg103233SHTI190421
ldquoJoint Statement on Contact Tracingrdquo 2020 April 19 httpscryptobriefingcomwp-contentuploads202004Joint-Statement-from-Researcherspdf
Kahn Gilmor Daniel 2020 ldquoPrinciples for Technology-Assisted Contact- Tracingrdquo White Paper American Civil Liberties Union httpswwwaclu orgreportaclu-white-paper-principles-technology-assisted-contact-tracing
Kim Max S 2020 ldquoSeoulrsquos Radical Experiment in Digital Contact Tracingrdquo The New Yorker April 17 2020 httpswwwnewyorkercomnewsnews-deskseouls-radical-experiment-in-digital-contact-tracing
Kim Nemo 2020 ldquolsquoMore Scary than Coronavirusrsquo South Korearsquos Health Alerts Expose Private Livesrdquo The Guardian March 6 2020 sec World news httpswwwtheguardiancomworld2020mar06more-scary- than-coronavirus-south-koreas-health-alerts-expose-private-lives
Kirzinger Ashley Liz Hamel Cailey Muntildeana Audrey Kearney and Mollyann Brodie 2020 ldquoKFF Health Tracking Poll ndash Late April 2020 Coronavirus Social Distancing and Contact Tracingrdquo Kaiser Family Foundation (blog) April 24 2020 httpswwwkfforgcoronavirus-covid-19issue-briefkff-health-tracking-poll-late-april-2020
Knobel Cory and Geoffrey C Bowker 2011 ldquoComputing Ethics Value in Designrdquo Communications of the ACM 54 (7) 26ndash28 httpsdoiorgdoi10114519657241965735
Lee Ramon Rosa R Cui Kathryn E Muessig Harsha Thirumurthy and Joseph D Tucker 2014 ldquoIncentivizing HIVSTI Testing A Systematic Review of the Literaturerdquo AIDS and Behavior 18 (5) 905ndash12 https doiorg101007s10461-013-0588-8
Leprince-Ringuet Daphne 2020 ldquoContact-Tracing Apps Why the NHS Said No to Apple and Googlersquos Planrdquo ZDNet April 28 2020 httpswwwzdnetcomarticlecontact-tracing-apps-why-the-nhs-said-no-to-apple-and- googles-plan
Lovejoy Ben 2020 ldquoNHS Contact Tracing App Hands-on as the Test Goes Liverdquo 9to5Mac (blog) May 7 2020 https9to5maccom20200507nhs-contact-tracing
Mahmood Sultan Khaled Hasan Michelle Colder Carras and Alain Labri-que 2020 ldquoGlobal Preparedness Against COVID-19 We Must Leverage the Power of Digital Healthrdquo JMIR Public Health and Surveillance 6 (2) e18980 httpsdoiorg10219618980
Mathews Simon C Michael J McShea Casey L Hanley Alan Ravitz Alain B Labrique and Adam B Cohen 2019 ldquoDigital Health A Path to
136 Works Cited
Validationrdquo Npj Digital Medicine 2 (1) 1ndash9 httpsdoiorg101038s41746-019-0111-3
Mello By Michelle M and C Jason Wang 2020 ldquoEthics and Governance for Digital Disease Surveillancerdquo Science May httpsdoiorg101126science abb9045
Mills Rodrigo Chris 2020 ldquoDigital Contact Tracing Is Becoming Available but Is It Effectiverdquo The Hill May 7 2020 httpsthehillcompolicytechnolo-gy496498-digital-contact-tracing-is-becoming-available-but-is-it-effective
MIT nd ldquoPrivate Kit Safe Paths Privacy-by-Design Covid19 Solutions Using GPS+Bluetooth for Citizens and Public Health Officialsrdquo Safepaths Accessed May 20 2020 httpssafepathsmitedu
ldquoMobile Cellular Subscriptions (per 100 People)rdquo 2018 World Telecommuni-cationICT Development Report International Telecommunication Union httpsdataworldbankorgindicatorITCELSETSP2
Moore Sarah Anne-Marie Tasseacute Adrian Thorogood Ingrid Winship Marsquon Zawati and Megan Doerr 2017 ldquoConsent Processes for Mobile App Mediated Research Systematic Reviewrdquo JMIR MHealth and UHealth 5 (8) e126 httpsdoiorg102196mhealth7014
Morse Jack 2020 ldquoNorth Dakota Launched a Contact-Tracing App Itrsquos Not Going Wellrdquo Mashable Accessed May 19 2020 httpsmashablecomarticlenorth-dakota-contact-tracing-app
Muller Robert T 2020 ldquoCOVID-19 Brings a Pandemic of Conspiracy Theo-riesrdquo Psychology Today April 24 2020 httpswwwpsychologytodaycomblogtalking-about-trauma202004covid-19-brings-pandemic-conspiracy- theories
Mulligan Stephen P and Chris D Linebaugh 2019 ldquoData Protection Law An Overviewrdquo R45631 Congressional Research Service httpscrsreports congressgovproductpdfRR45631
National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research 1979 ldquoThe Belmont Report Ethical Principles and Guidelines for the Protection of Human Subjects of Researchrdquo US Dept of Health amp Human Services Office of Human Research Protections httpswwwhhsgovohrpregulations-and-policybelmont-reportread-the-belmont-reportindexhtml
NDDoH 2020 ldquoNorth Dakota Launches Care19 App to Combat COVID-19rdquo North Dakota Department of Health April 7 2020 httpswwwhealth ndgovnewsnorth-dakota-launches-care19-app-combat-covid-19
NYC DOH 2020 ldquoAge-Adjusted Rates of Lab Confirmed COVID-19 Non-hospitalized Cases Estimated Non-Fatal Hospitalized Cases and Persons Known to Have Died per 100000 by RaceEthnicity Grouprdquo New York
Works Cited 137
City Department of Health httpswww1nycgovassetsdohdownloadspdfimmcovid-19-deaths-race-ethnicity-04242020-1pdf
OrsquoRielly Michael 2018 ldquoFCC Regulatory Free Arenardquo Federal Communica-tions Commission June 1 2018 httpswwwfccgovnews-eventsblog 20180601fcc-regulatory-free-arena
OSHA 2020 ldquoGuidance on Preparing Workplaces for COVID-19rdquo OSHA 3990-03 2020 Department of Labor Occupational Safety and Health Act of 1970 httpswwwoshagovPublicationsOSHA3990pdf
Pallmann Philip Alun W Bedding Babak Choodari-Oskooei Munyaradzi Dimairo Laura Flight Lisa V Hampson Jane Holmes et al 2018 ldquoAdaptive Designs in Clinical Trials Why Use Them and How to Run and Report Themrdquo BMC Medicine 16 (1) 29 httpsdoiorg101186s12916-018-1017-7
Pew Research Center 2017 ldquoUS Muslims Concerned About Their Place in Society but Continue to Believe in the American Dreamrdquo Religion amp Public Life Pew Research Center httpswwwpewforumorg20170726findings-from-pew-research-centers-2017-survey-of-us-muslims
mdashmdashmdash 2020 ldquoDemographics of Mobile Device Ownership and Adoption in the United Statesrdquo Internet amp Technology Pew Research Center Accessed May 19 2020 httpswwwpewresearchorginternetfact-sheetmobile
PHLS 2002 ldquoPrinciples of the Ethical Practice of Public Healthrdquo Public Health Leadership Society httpswwwaphaorg-mediafilespdfmembergroupsethicsethics_brochureashx
PIH 2020a ldquoPart I Testing Contact Tracing and Community Management of COVID-19rdquo PIH Guide | COVID-19 STOP COVID Partners in Health httpswwwpihorgsitesdefaultfiles2020-04PIH_Guide_COVID_Part_I_Testing_Tracing_Community_Managment_4_4pdf
mdashmdashmdash 2020b ldquoBox It Inrdquo PIH Guide | COVID-19 Partners in Health httpspreventepidemicsorgcovid19resourcesbox-it-in
Resolve to Save Lives 2020 ldquoCOVID-19 Contact Tracing Playbookrdquo Vital Strategies httpscontacttracingplaybookresolvetosavelivesorg
Reston Maeve Kristina Sgueglia and Cheri Mossburg 2020 ldquoGovernors on East and West Coasts Form Pacts to Decide When to Reopen Economiesrdquo CNN Politics April 13 2020 httpswwwcnncom20200413politicsstates-band-together-reopening-plansindexhtml
Rithalia Amber Catriona McDaid Sara Suekarran Lindsey Myers and Amanda Sowden 2009 ldquoImpact of Presumed Consent for Organ Donation on Donation Rates A Systematic Reviewrdquo BMJ 338 (January) httpsdoiorg101136bmja3162
Rodrigues Rafaela Alina Husain Amanda Couture-Carron Leslye E Orloff
138 Works Cited
and Nawal H Ammar 2018 ldquoPromoting Access to Justice for Immigrant and Limited English Proficient Crime Victims in an Age of Increased Immi-gration Enforcement Initial Report from a 2017 National Surveyrdquo Wash-ington DC National Immigrant Womenrsquos Advocacy Project American University Washington College of Law httpniwaplibrarywclamericaneduwp-contentuploadsImmigrant-Access-to-Justice-National-Reportpdf
Simpson Erin and Adam Conner 2020 ldquoDigital Contact Tracing to Contain the Coronavirusrdquo Center for American Progress httpswww americanprogressorgissuestechnology-policynews20200422483521digital-contact-tracing-contain-coronavirus
Singer Eleanor and Cong Ye 2013 ldquoThe Use and Effects of Incentives in Surveysrdquo The ANNALS of the American Academy of Political and Social Science 645 (1) 112ndash41 httpsdoiorg1011770002716212458082
Thornton Rebecca L 2008 ldquoThe Demand for and Impact of Learning HIV Statusrdquo The American Economic Review 98 (5) 1829ndash63 httpsdoi org101257aer9851829
US DOE 2009 ldquoState Regulation of Private Schoolsrdquo US Dept of Education Office of Non-Public Education httpswww2edgovadminscommchoiceregprivschlregprivschlpdf
Valentino-DeVries Jennifer 2020 ldquoCellphone Carriers Face $200 Million Fine for Not Protecting Location Datardquo The New York Times February 28 2020 sec Technology httpswwwnytimescom20200228technologyfcc-cellphones-location-data-fineshtml
Valentino-DeVries Jennifer Natasha Singer and Aaron Krolik 2020 ldquoA Scramble for Virus Apps That Do No Harmrdquo The New York Times April 29 2020 sec Business httpswwwnytimescom20200429businesscoronavirus-cellphone-apps-contact-tracinghtml
Washington PostndashUMD ldquoWashington Post-University of Maryland National Pollrdquo 2020 Washington Post April 21 2020 httpswwwwashingtonpostcomcontextwashington-post-university-of-maryland-national-poll-april-21-26-20203583b4e9-66be-4ed6-a457-f6630a550ddf
Watson Crystal Anita Cicero James Blumenstock and Michael Fraser 2020 ldquoA National Plan to Enable Comprehensive COVID-19 Case Finding and Contact Tracing in the USrdquo Johns Hopkins Bloomberg School of Public Health Center for Health Security and the Association of State and Terri-torial Health Officials httpswwwcenterforhealthsecurityorgour-workpubs_archivepubs-pdfs2020200410-national-plan-to-contact-tracingpdf
Works Cited 139
WHO 2017 ldquoWHO Guidelines on Ethical Issues in Public Health Surveillancerdquo World Health Organization httpswwwwhointethicspublicationspublic-health-surveillanceen
mdashmdashmdash 2020 ldquoCoronavirus Disease 2019 (COVID-19)rdquo Situation Report 73 World Health Organization httpswwwwhointdocsdefault-sourcecoro-navirusesituation-reports20200402-sitrep-73-covid-19pdf
- _iy90yrtqgtxz
- _GoBack
- Acknowledgments
- Preface
- Lead Authors and Contributors
- Acronyms and Abbreviations
- Summary
-
- Introduction
- DCTT Features Functions and Potential Applications
- Summary of Recommendations
- Summary of Analysis
-
- Introduction
-
- Guiding Principles for the Use of Digital Public Health Technologies for Pandemic Response
-
- Public Health Perspective
-
- Types of Information Collected through Contact Tracing
- How Contact Tracing Information Informs Public Health Action
- Characteristics That Make Data Useful to Public Health for Reducing Disease Transmission
-
- Digital Technology and Contact Tracing
-
- Characteristics of SARS-CoV-2 Relevant to Candidate Digital Solutions
- Previously Existing Contact Tracing Technologies
- Introduction of Novel Digital Contact Tracing Technologies
- Relevant Differences between Manual and Digital Contact Tracing
-
- Ethics of Designing and Using DCTT
-
- Justifying the Use of DCTT Systems
- Monitoring and Evaluating Technologies to Inform Policy and Practice
- Public Trust and Public Attitudes
- Designing Flexible Technology to Maximize Public Health Utility While Respecting Other Values
- Policy Positions to Advance Widespread Use of Digital Contact Tracing Technologies
- Disclosure and AuthorizationConsent
- Promoting Equity and Fairness
- Instituting Transparent Governance and Oversight
-
- Legal Considerations
-
- Data Privacy and Data Security Laws
- Health Information Privacy
- Labor and Employment Privacy Rights
- Constitutional Privacy Rights
- Consent
- Anti-discrimination and Individual Freedom Laws
-
- Recommendations
-
- Public Health
- Ethics
- Legislative
-
- Resources
-
- US Government Response
- Other Governmental and Nongovernmental Organizations
- Digital Contact Tracing Experiences from Other Countries
- Specific Digital ProductsApps
- Polling
- Popular Press
- Commentaries
- Academic Literature
-
- Works Cited
-
Contents
Lead Authors and Contributors vii
Preface ix
Acknowledgments xi
Acronyms and Abbreviations xiii
Summary 1
Introduction 1
DCTT Features Functions and Potential Applications 3
Summary of Recommendations 9
Summary of Analysis 13
Introduction 23
Guiding Principles for the Use of Digital Public Health
Technologies for Pandemic Response 25
1 Public Health Perspective 29
Types of Information Collected through Contact Tracing 29
How Contact Tracing Information Informs Public Health Action 31
Characteristics That Make Data Useful to
Public Health for Reducing Disease Transmission 32
2 Digital Technology and Contact Tracing 35
Characteristics of SARS-CoV-2 Relevant to Candidate Digital Solutions 35
Previously Existing Contact Tracing Technologies 36
Introduction of Novel Digital Contact Tracing Technologies 37
Relevant Differences between Manual and Digital Contact Tracing 41
vi Contents
3 Ethics of Designing and Using DCTT 43
Justifying the Use of DCTT Systems 44
Monitoring and Evaluating Technologies to Inform Policy and Practice 45
Public Trust and Public Attitudes 48
Designing Flexible Technology to Maximize Public Health Utility
While Respecting Other Values 50
Policy Positions to Advance Widespread Use of Digital Contact
Tracing Technologies 59
Disclosure and AuthorizationConsent 63
Promoting Equity and Fairness 69
Instituting Transparent Governance and Oversight 72
4 Legal Considerations 75
Data Privacy and Data Security Laws 77
Health Information Privacy 82
Labor and Employment Privacy Rights 84
Constitutional Privacy Rights 86
Consent 93
Anti-discrimination and Individual Freedom Laws 94
5 Recommendations 97
Public Health 97
Ethics 98
Legislative 102
Resources 103
US Government Response 103
Other Governmental and Nongovernmental Organizations 106
Digital Contact Tracing Experiences from Other Countries 112
Specific Digital ProductsApps 115
Polling 117
Popular Press 119
Commentaries 120
Academic Literature 121
Works Cited 127
Lead Authors and Contributors
vii
Lead Authors
Joseph Ali JD Assistant Professor Department of International Health Johns Hopkins Bloomberg School of Public Health Core Faculty amp Associate Director for Global Programs Johns Hopkins Berman Institute of Bioethics
Anne Barnhill PhD Core Faculty amp Research Scholar Johns Hopkins Berman Institute of Bioethics
Anita Cicero JD Deputy Director Johns Hopkins Center for Health Security Visiting Faculty Johns Hopkins Bloomberg School of Public Health
Katelyn Esmonde PhD Hecht-Levi Postdoctoral Fellow Johns Hopkins Berman Institute of Bioethics
Amelia Hood MA Research Program Coordinator Johns Hopkins Berman Insti-tute of Bioethics
Brian Hutler PhD JD Hecht-Levi Postdoctoral Fellow Johns Hopkins Berman Institute of Bioethics
Jeffrey Kahn PhD MPH Andreas C Dracopoulos Director Johns Hopkins Ber-man Institute of Bioethics
Alan Regenberg MBE Director of Outreach amp Research Support Associate Fac-ulty Johns Hopkins Berman Institute of Bioethics
Crystal Watson DrPH MPH Senior Scholar Johns Hopkins Center for Health Security Assistant Professor Department of Environmental Health and Engi-neering Johns Hopkins Bloomberg School of Public Health
Matthew Watson Senior Analyst Johns Hopkins Center for Health Security Senior Research Associate Department of Environmental Health and Engi-neering Johns Hopkins Bloomberg School of Public Health
Other Contributors
Robert Califf MD MACC Head of Clinical Policy and Strategy Verily and Goo-gle Health
Ruth Faden PhD MPH Philip Franklin Wagley Professor of Biomedical Ethics amp Founder Johns Hopkins Berman Institute of Bioethics
viii Lead Authors and Contributors
Divya Hosangadi MSPH Senior Analyst Johns Hopkins Center for Health Secu-rity Research Associate Department of Environmental Health and Engi-neering Johns Hopkins Bloomberg School of Public Health
Nancy Kass ScD Deputy Director for Public Health amp Phoebe R Berman Pro-fessor of Bioethics and Public Health Johns Hopkins Berman Institute of Bioethics
Alain Labrique PhD MHS MS Professor Department of International Health Johns Hopkins Bloomberg School of Public Health Director Johns Hopkins Global Health Initiative
Deven McGraw JD MPH LLM Chief Regulatory Officer CiitizenMichelle Mello JD PhD Professor of Law Stanford Law School Professor of
Health Research and Policy Stanford University School of MedicineMichael Parker BEd (Hons) MA PhD Director Wellcome Centre for Ethics and
Humanities Ethox Centre University of OxfordStephen Ruckman JD MSc MAR Senior Advisor to the President for Policy
Office of the President Johns Hopkins UniversityLainie Rutkow JD MPH PhD Senior Advisor to the President for National Cap-
ital Academic Strategy Office of the President Johns Hopkins UniversityJosh Sharfstein MD Vice Dean for Public Health Practice and Community
Engagement Professor of the Practice Johns Hopkins Bloomberg School of Public Health
Jeremy Sugarman MD MPH MA Deputy Director for Medicine Harvey M Meyerhoff Professor of Bioethics and Medicine Johns Hopkins Berman Institute of Bioethics Department of Medicine Johns Hopkins School of Medicine and Department of Health Policy and Management Johns Hop-kins Bloomberg School of Public Health
Eric Toner MD Senior Scholar Johns Hopkins Center for Health Security Senior Scientist Department of Environmental Health and Engineering Johns Hop-kins Bloomberg School of Public Health
Marc Trotochaud MSPH Analyst Johns Hopkins Center for Health Security Research Associate Department of Environmental Health and Engineering Johns Hopkins Bloomberg School of Public Health
Effy Vayena PhD Professor Health Ethics amp Policy Lab Department of Health Sciences amp Technology ETH Zurich
Tal Zarsky JSD LLM LLB Professor of Law University of Haifa Faculty of Law Visiting Scholar University of Pennsylvania Law School (2019ndash2020)
Preface
ix
Digital technologies are being developed and promoted to support the public health response to the COVID-19 pandemic with discussion and implementation planning in the United States by localities states institu-tions and employers Key decision makers and stakeholdersmdashincluding government officials institutional leaders employers digital technology developers and the publicmdashrequire clear and well-supported guidance to inform the deployment and use of these technologies as well as of the data they collect store and share While technology-based approaches are currently unable to provide solutions on their own experiences in other countries indicate that they could be used successfully in conjunc-tion with traditional and novel public health methods
This report reflects a rapid research and expert consensus group ef-fort led by the Berman Institute of Bioethics and the Center for Health Se-curity at Johns Hopkins University It draws on experts from both inside and outside Johns Hopkins in bioethics health security public health technology development engineering public policy and law The report highlights issues that must be addressed and provides recommendations for the use of digital technologies as part of contact tracing
The analysis offered here is focused on answering the following questions
bull Can digital contact tracing technologies (DCTT) be effective as part of public health responses to the pandemic and if so to what degree for which specific types of functions with what confidence and with what requirements
bull How can these technologies serve the interests of public health while respecting other individual and collective interests such as ensuring equitable distribution of benefits and burdens and limit-ing infringement on privacy and other civil liberties
x Preface
bull What are the ethical legal policy and governance guardrails cur-rently in place around such technologies and what else is needed
bull What additional guardrails are required to ensure that the goals of public health in using these technologies are achievable in ways that are ethically and legally sound
To answer these questions the report examines some core aspects of dig-ital technologies applied to contact tracing focusing on
bull the value of and basic methods for traditional public health surveil-lance and contact tracing
bull candidate technological products to enhance public health surveil-lance and contact tracing how they work and their comparative value for public health
bull core ethical legal and governance considerations and how they relate to relevant features of candidate technological solutions and
bull what is needed to move forward responsibly with the use of digital technology in support of public health surveillance acknowledg-ing gaps in our current understanding
The project involved in-depth analysis by a dedicated team of faculty postdoctoral fellows and research staff working over the course of only a few weeks but with great intensity drafting a report in collaboration with 26 total contributors writing commenting and revising through multiple drafts with the penultimate draft ldquopressure-testedrdquo by review and discus-sion at a virtual workshop of invited experts and stakeholders held on May 13 2020 and the final version completed on May 21 2020 The report and analysis builds on the excellent work of others in some parts of this territory while focusing on the gaps in analysis and areas that have not been sufficiently addressed The goal is to offer comprehensive guid-ance to relevant stakeholders to advance public health response during the COVID-19 pandemic Given the rapidly evolving territory into which DCTT is being introduced this report will by necessity be something of a living document updated as often as information dictates in order to continue to offer leading-edge analysis and guidance Versions will be noted in the digital and print editions
Acknowledgments
xi
Efforts like this project require teams and even small armies to be carried out successfully and this was no exception except that it was many fewer people working many more hours than could reasonably be expected of them From the initial kernel of an idea to the publication of this report in book form this project took just over a month total That seems impossi-ble even as I know it is accurate and it speaks to the incredible commit-ment hard work research skills and analytic acumen of our colleagues at Johns Hopkinsmdashthe core team are deservedly listed as lead authors of this report
None of this would have been possible without the supportmdashmoral and financialmdashand encouragement of Johns Hopkins University Presi-dent Ronald J Daniels who was the first to suggest the idea to me of taking on this topic He provided not only support and encouragement but the imprimatur of his office including help guidance and counsel from Prof Lainie Rutkow senior advisor to the president Lainie played a more integral role than that description captures reflected in part by her inclusion among the reportrsquos contributors but she deserves special acknowledgement for shepherding us through to the end
I mentioned that this was a team effort and every team requires an effective leader My colleague Prof Joseph Ali stepped into that role as we undertook the project and then he worked seven days a week along with the rest of the core research and writing team always unfailingly positive and deeply engaged in the work He along with Prof Anne Barnhill Alan Regenberg Amelia Hood and Drs Katelyn Esmonde Brian Hutler and Crystal Watson all deserve special thanks for doing so much in so little time all while working under the grinding social distancing restrictions of the 2020 pandemic That work was supported by Arnold amp Porter Kaye Scholer LLP with legal research and other assistancemdasha huge thanks to
xii Acknowledgments
them Finally the 16 contributing authors were incredibly generous with their time energy and insights all on ridiculously tight timelines and never a complaint or objection
The project benefited greatly from a number of experts who provided written feedback on drafts and who attended the virtual workshop to test our recommendations including Miles Stewart Rob Nichols Smisha Aagarwal Karl Steiner Anupam Joshi Charles Scheeler Ford Rowan and Jay Wagley
Last the fact that this report appears in published book form by Johns Hopkins University Press is another minor miracle from manu-script to printed book in under a week Thanks to JHUP Director Barbara Kline Pope and her team for being willing to take on the challenge and for the incredible focused effort it required
my heartfelt thanks andappreciation to you all
Jeffrey Kahn
Acronyms and Abbreviations
xiii
ADA Americans with Disabilities ActBLE Bluetooth Low EnergyCalOPPA California Online Privacy Protection ActCBP Customs and Border ProtectionCCPA California Consumer Privacy ActCDC Centers for Disease Control and PreventionCLOUD Act Clarifying Lawful Overseas Use of Data ActCOPPA Childrenrsquos Online Privacy Protection ActCOV+ confirmed positive SARS-CoV-2 test resultCOVID-19 coronavirus disease 2019CPNI customer proprietary network informationCSLI cell-site location informationDCTT digital contact tracing technology and closely related
digital health productsECPA Electronic Communications Privacy ActEEOC Equal Employment Opportunity CommissionE-SIGN Electronic Signatures in Global and National
Commerce ActEU European UnionFCC Federal Communications CommissionFTC Federal Trade CommissionFTCA Federal Trade Commission ActGIS geographic information systemGPS global positioning systemHHS US Department of Health and Human ServicesHIPAA Health Insurance Portability and Accountability ActHIV human immunodeficiency virusICU intensive care unit
xiv Acronyms and Abbreviations
IRB institutional review boardJHU Johns Hopkins UniversityOCR Office for Civil Rights US Department of Health amp
Human ServicesOSHA Occupational Safety and Health AdministrationPHI protected health informationPII personally identifiable informationPPE personal protective equipmentPPPT privacy-preserving proximity trackingQR Code quick response codeRFRA Religious Freedom Restoration ActRLUIPA Religious Land Use and Institutionalized Persons ActSARS-CoV-2 severe acute respiratory syndrome coronavirus 2SCA Stored Communications ActSTI sexually transmitted infection
DIGITAL CONTACT TRACING FOR PANDEMIC RESPONSE
11
Summary
Introduction
Public health professionals around the world are working tirelessly to respond to the COVID-19 pandemic using tried-and-tested public health methods for infectious disease surveillance and control These traditional methods are essential to the global COVID-19 response To complement these actions and potentially augment the speed and efficacy of the pub-lic health workforce digital technologies are being harnessed Given the scale of the pandemic significant efforts are being undertaken to develop and leverage public-facing and health-system-supportive technology solu-tions including smartphone apps and other digital tools that may aid public health surveillance and contact tracing
Digital contact tracing technology and closely related digital health products (together DCTT) have been used in several countries as part of broader disease surveillance and containment strategies In the United States DCTT has been proposed as an integral part of some plans to ldquoreopenrdquo the country (Allen et al 2020 Hart et al 2020 Simpson and Conner 2020) It is almost certain that these and related technologies will become part of not only the COVID-19 response but also the larger toolbox for future public health communicable disease prevention and control
These technologies have significant promise They also raise import-ant ethical legal and governance challenges that require comprehensive analysis in order to support decision-making Government officials pub-lic health leaders leaders of institutions employers digital technology de-velopers and the public all must be adequately informed in order to make
2 Digital Contact Tracing for Pandemic Response
responsible choices Johns Hopkins University recognized the importance of helping to guide this process It organized an expert group with mem-bers from inside and outside of Hopkins and led by its Berman Institute of Bioethics in collaboration with the Center for Health Security Its charge was to examine the ethics law policy and public health implications of using digital technologies as part of pandemic response and to develop guidance including a framework and actionable recommendations for governmental and institutional decision makers
Overall this expert group urges a stepwise approach that prioritizes align-ment of technology with public health needs and public values building choice into design architecture and capturing real-world results and impacts to allow adjustments as required Further we urge an approach that recog-nizes that there are complicated issues to resolve for governments insti-tutions and businesses and that introduction of DCTT must include public engagement and ongoing assessments to improve both performance and adoption
Specific recommendations include the following
bull There is no ldquoone size fits allrdquo approach to DCTT Technology design should not be static but it should be capable of evolving depending upon local conditions new evidence and changing preferences and priorities
bull Technology companies alone should not control the terms condi-tions or capabilities of DCTT nor should they presume to know what is acceptable to members of the public
bull DCTT should be designed to have a base set of features that pro-tect privacy with layers of additional capabilities that users may choose to activate An initial default should be that user location data are not shared but users should be provided with easy mech-anisms and prompts to allow for opting-in to this capability with encouragement to the public if it is shown to be critical to achieving public health goals
bull Data collected through DCTT should be made available to public health professionals and to researchers in de-identified form to support population-level epidemiologic analyses
Summary 3
bull Those who authorize use of DCTT within a particular jurisdic-tion or institution should continuously and systematically monitor the technologyrsquos performance in that context This should include monitoring for effectiveness and benefit monitoring for harms and monitoring for the fair distribution of both benefits and harms
bull Governments should not require mandatory use of DCTT given uncertainty about potential burdens and benefits Additional tech-nology user and real-world testing is needed
Through in-depth analysis and recommendations this report seeks to guide decision-making and enhance understanding of
bull the value of and basic methods for traditional public health surveil-lance and contact tracing
bull candidate technological products to enhance public health surveil-lance and contact tracing and their comparative value for public health
bull core ethical legal policy and governance considerations and how they relate to relevant features of candidate technological solutions and
bull what is needed to move forward responsibly with the use of digital technology in support of public health surveillance acknowledg-ing gaps in our current understanding
The full set of recommendations are intended to (1) support effective and informed adoption of DCTT (2) encourage design of flexible technol-ogies that maximize public health utility while respecting other values (3) establish meaningful processes for user disclosure and authorization (consent) (4) promote equity and fairness in the uses of DCTT and (5) foster transparent governance and oversight
DCTT Features Functions and Potential Applications
Digital contact tracing technologies and platforms can be roughly catego-rized into three broad approaches along a spectrum of potential policies and methods a maximal approach (typified by the South Korean govern-
4 Digital Contact Tracing for Pandemic Response
mentrsquos centralized and triangulated data collection (M S Kim 2020)) a minimal approach (typified by the AppleGoogle decentralized priva-cy-preserving proximity tracking (PPPT) and contact notification (Apple and Google nd)) and a diverse range of middle-ground approaches that aim to augment manual contact tracing with the collection of digital data that can be shared with public health authorities
Minimal approaches such as the AppleGoogle PPPT use Bluetooth Low Energy (BLE) ldquohandshakesrdquo that record close contact between mo-bile phone users but do not register the location in which the contact hap-pened In most architectures these proximity data are stored in the usersrsquo phones as anonymized ldquobeaconsrdquo that cannot be used to re-identify the users directly If a user with a PPPT app installed on their phone tests pos-itive and enters test results into their app those who have been in contact with them can be notified by the app This ldquoexposure notificationrdquo can be automatic or at the discretion of the COV+ person depending on the app design If notified a user who has been in contact with a COV+ individ-ual would receive a push notification alerting them to possible exposure (which may be timestamped) but with no other identifying information
The most prevalent middle-ground approach in the US context in-volves the collection and storage of personal datamdashincluding identifying information and location datamdashon the userrsquos phone These decentralized but personally identifiable data can then be voluntarily shared with pub-lic health officials if the user tests positive for SARS-CoV-2 (severe acute respiratory syndrome coronavirus 2) For example a team at the Mas-sachusetts Institute of Technology (MIT) has developed an app called Private Kit Safe Paths (MIT nd) that stores usersrsquo location data on their phone for 28 days If a user tests positive she can voluntarily upload her location data to a website that is accessible only to public health officials Officials can then analyze these personally identifiable data and subse-quently broadcast redacted and de-identified data to other users Healthy users would have access to these redacted location data of COV+ users but their own data would not leave their phones At a minimum the storage of user location data can function as a ldquomemory aidrdquo if the user tests positive but releasing the data to public health authorities may help to analyze the spread of SARS-CoV-2 and alert individuals or groups that have been in contact with COV+ patients
Summary 5
The US Centers for Disease Control and Prevention (CDC) has pub-lished preliminary criteria for evaluating capabilities and attributes of DCTT (CDC 2020e) These and other resources suggest that a compre-hensive assessment of DCTT and its potential to advance the publicrsquos health will require careful consideration of numerous interconnected fac-tors that interact in complex ways and must be navigated within the chal-lenging contexts of uncertainty and urgent need (Figure 1) These include
bull scientific and epidemiological understanding of SARS-CoV-2 transmission and infection
bull public health needs for combating the outbreak
bull technological capabilities of DCTT
bull performance of DCTT applications
bull ethical values and principles
bull characteristics of public adoption and acceptance and
bull legal issues and landscape
FIGURE 1 Interrelating Factors That Frame Responsible Development of Digital Contact
Tracing Technology
6 Digital Contact Tracing for Pandemic Response
The primary objectives for use of DCTT during the COVID-19 pandemic must be to reduce illness and death and facilitate public health efforts to reduce transmission of the virus These objectives fall under a broader overall goal of contributing to societal well-being during the pandemic It is not yet known whether and how much DCTT can contribute to these primary objectives nor whether it will be able to contribute without gen-erating new burdens or even harms such as incorrect warnings or ldquonoiserdquo that detract from the work of manual contact tracing
The process of identifying acceptable technology designs and uses is complex given the interplay among the factors Our analysis reveals that there is no ldquoone size fits allrdquo approach to DCTT There is variability across the United States with respect to SARS-CoV-2 prevalence and in-fection rates public health capacity public attitudes toward DCTT and acceptability of various potential features Moreover our understanding of SARS-CoV-2 and DCTT is evolving public health response needs and capabilities are changing and public attitudes are shifting Different tech-nologies used in different ways may be appropriate to achieve slightly dif-ferent public health goals in different localities and at different points in the pandemic A tiered and phased approach to technology development should be facilitated by law and policy prioritizing underlying interoper-ability while permitting user choices now and for the future
Given the complexity of the terrain as a first step those developing or considering widespread use of DCTT as part of pandemic response should be guided by the following principles and related actions (see box) These principles are meant to apply to DCTT as well as other dig-ital technologies used in novel ways during pandemic response
These principles make clear that in order to maximize the public good from use of DCTT public health needs and technological capabili-ties must be carefully aligned Government officials public health leaders leaders of other institutions employers digital technology developers and the public are all key stakeholders that must be informed and en-gaged in order to enable the most successful and ethically acceptable uses of DCTT
Guiding Principles for the Use of Digital Public Health Technologies for Pandemic Response
Transparency and public engagement are essential to an inclusive digital public health response
bull Government public health and digital technology leaders must engage
effectively with the public and other stakeholders to communicate the util-
ity importance oversight and limitations of relevant digital technologies
including their implications for individualsrsquo privacy and civil liberties
bull Transparency at all levels is essential for maintaining public trust and
confidence
bull To the extent possible digital public health responses should reflect the
range of values that are important to individuals including advancing the
health and well-being of the community as a whole
bull Decision makers should recognize the sacrifices that some people may be
willing to make during a pandemic in order to advance public health goals
Acceptance by some of particular monitoring capabilities should not be read
as a willingness to extend these methods to other problems or uses
Digital public health responses must represent the least infringement of civil liberties necessary to accomplish the public health goals
bull If preferred digital public health strategies infringe on privacy and other civil
liberties the infringements must be sufficiently justified by the circum-
stances of the pandemic offset by ample anticipated public benefit and
considered relative to infringements associated with other possible strate-
gies such as mass physical distancing
bull Only those data that are necessary and relevant for the stated public health
purposes should be collected Identifiable data should be stored in a secure
manner and only for the period of time that the public health purposes
require
bull Adopted technologies should not be used in ways that subject communities
to discrimination or surveillance for nonndashpublic health reasons
bull Respect for individual autonomy requires that users are sufficiently informed
of the public health goals of the technology and the extent to which those
goals are being met
Use of digital public health technologies and data must be guided by best available evidence
bull Decisions to deploy digital public health technologies should be based on a
careful assessment of the uses and limitations of any proposed technology
taking into account the best available evidence
bull Those who deploy digital public health technologies should continuously and
systematically monitor their performance as well as any evidence that is being
generated in other contexts about the selected technological solution and about
other competing technologies
bull Unintended consequencesmdashincluding those that might impact public health
goals core values and interests of the public and unfair advantage or disadvan-
tagemdashshould be carefully monitored and addressed as necessary
Responsible use of digital public health technology requires meaningful governance and accountability
bull Systems of governance must be trustworthy and well informed They must be
reviewed and adjusted as circumstances and evidence change or as unintended
effects are identified
bull Trusted representatives who are capable of developing and implementing uni-
form and fair standards for adopting and utilizing underlying digital technology
must be identified
bull Understandable transparent and publicly accessible rules must guide the
collection access control use storage and combination of data by government
authorities public and private institutions and other parties such as public
health researchers
bull Oversight accountability and consequences for abuse or misuse of these data
must be explicit and enforceable
The deployment of digital public health technology must be rooted in a commitment to equity
bull Digital public health technologies should be deployed in a manner that does
not propagate preexisting patterns of unfair disadvantage or further distribute
harms and risks unfairly throughout the population
bull To the extent possible digital public health technologies should be designed to
rectify existing inequities
bull Oversight mechanisms must be in place to ensure that the improved public
health outcomes are equitable and to detect and correct any unforeseen resul-
tant injustices attributable to the technology or that can be addressed using the
technology
bull The incentives and disincentives for adopting new technology must be equita-
ble not exploitative and aligned with effective use of the technology
bull Disparity-driven technology gaps should be explicitly recognized To the extent
possible provisions should be made to address the digital divide
Summary 9
Summary of Recommendations
The guidance document makes a number of recommendations related to (1) supporting effective and informed adoption of DCTT (2) designing flexible technologies to maximize public health utility while respecting other values (3) establishing meaningful processes for user disclosure and authorizationconsent (4) promoting equity and fairness in application of DCTT and (5) instituting transparent governance and oversight Here we provide a summary of recommendations
Supporting Effective and Informed Adoption
bull Those who authorize use of DCTT within a particular jurisdic-tion or institution should continuously and systematically monitor the technologyrsquos performance in that context This should include monitoring for effectiveness and benefit monitoring for harms and monitoring for the fair distribution of both benefits and harms They should also monitor evidence that is being generated in other contexts about their selected technological solution and about other competing technologies
bull Data collected through DCTT should be made available to public health professionals and to researchers in de-identified form to support population-level epidemiologic analysis
bull Data should be available to users that would permit them to further investigate their personal risk with public health officials or other health workers to add a layer of protection against unnecessary quarantine
bull Technologies or apps may produce some false negatives or false positives but they should be accurate enough that public health authorities feel confident that they support and donrsquot detract from contact tracing efforts
bull Trusted leaders should be enlisted to communicate effectively with the public about DCTT and encourage its use should the technol-ogy demonstrate some potential The limits of knowledge regard-ing effectiveness should also be explained along with what will be done to improve technological capabilities as understanding evolves
10 Digital Contact Tracing for Pandemic Response
bull Incentives can be a useful complement to encouragements how-ever any incentives for users to install and use DCTT must be equitable should not be coercive and should align with effective use of the technology
bull DCTT use should not be mandated at this time given uncertainty about potential harms and benefits Additional technology user and real-world testing is needed
Designing Flexible Technology to Maximize
Public Health Utility While Respecting Other Values
bull Technology companies should not alone control the terms condi-tions or capabilities of DCTT nor should they presume to know what is acceptable to members of the public
bull A ldquovalues in designrdquo approach to development of DCTT should be adopted (Flanagan Howe and Nissenbaum 2008 Knobel and Bowker 2011) Robust public and user engagement activities should be pursued to identify and incorporate to the extent pos-sible a range of values into the design of the technology These values may include privacy but also autonomy efficiency equity or others Technology design should reflect an appropriate balance and prioritization of identified values
bull Technology design should not be static but should be capable of evolving depending upon local conditions new evidence and changing preferences and priorities
bull DCTT should be designed to have a base set of features that pro-tect privacy with layers of additional capabilities that users may choose to activate An initial default should be that user location data are not shared but users should be provided with easy mech-anisms and prompts to allow for opting-in to this capability with encouragement to the public if it is shown to be critical to achieving public health goals
Establishing Meaningful Processes for
User Disclosure and Authorization (Consent)
bull A clear and concise module consisting of basic disclosure and vol-untary authorization should be developed to accompany DCTT
Summary 11
This module should not take the form of ldquoclickwraprdquo terms of service or end-user agreements but rather provide only essential information necessary for an individual to make a decision More detailed disclosures (such as FAQs in plain language) should be made easily accessible to those who wish to learn more with no hidden surprises
bull An opt-in approach to authorization should be instituted to ac-company initial DCTT rollout The feasibility and value of opt-out approaches should continue to be evaluated informed by what is technologically possible what local assessments of benefits and harms of the technology reveal over time and our evolving un-derstanding of the degree to which an opt-out approach is likely to increase or decrease utilization among different populations Opt-out approaches should not be precluded
Promoting Equity and Fairness in Application of DCTT
bull States localities and institutions that recommend widespread use of DCTT should provide technology (eg mobile phones Blue-tooth devices) and free data packages to those who desire but lack access to these devices
bull If there are lower rates of adoption of DCTT systems in some iden-tifiable communities public health authorities should find ways to compensate For example directing more non-DCTT resources and efforts toward those communities to meet specific needs that are elsewhere being supported by technology
bull If maps are generated based on DCTT to provide the public with the locations that COV+ individuals have visited steps must be taken to minimize the stigma and potential financial losses that could result from a location being identified as a hotspot
Instituting Transparent Governance and Oversight
bull Digital surveillance oversight committees should be established expeditiously with diverse and qualified membership to provide ethical and regulatory review prior to and concurrent with wide-spread use of a DCTT system
bull Understandable and publicly accessible rules must guide the col-
12 Digital Contact Tracing for Pandemic Response
lection access control use storage and combination of data by government authorities public and private institutions and other parties such as public health researchers
bull Only those data that are necessary and relevant for the public health response to COVID-19 should be collected and used
bull Identifiable data should be kept only for the period of time needed for the public health response to COVID-19
bull Identifiable data collected as part of this response should not be shared with anyone other than the relevant public health authorities without additional specific informed consent of individual users
bull Before a government or institution adopts a digital contact tracing program they should state the conditions under which the digital contact tracing program will be terminated
bull Future use of DCTT to advance public health or other efforts (eg use in seasonal flu surveillance) would require independent justification DCTT designed for public health use should not be used by law or immigration enforcement
bull The principles offered in this guidance document apply both during and following the COVID-19 pandemic
Legislative Recommendations
bull The United States Congress should enact new legislation specif-ically tailored to facilitate the use of DCTT as part of the public health response to COVID-19 while also protecting user privacy and ensuring data security
bull Congress should require DCTT developers to disclose to users in clear language the nature of the information that would be col-lected how it would be collected how it would be stored and for what purposes it may be used
bull While the rollout of DCTT should initially employ an opt-in au-thorization approach the feasibility acceptability and value of opt-out approaches should continue to be evaluated As such opt-out approaches to consent should not be precluded by legislation
Summary 13
bull Congress should prohibit the commercial use of data collected for COVID-19 response by DCTT
bull Congress should prohibit discrimination on the basis of data col-lected by DCTT
bull If Congress is unable to enact suitable legislation state legislatures should work toward enacting similar laws for their jurisdictions A ldquomodelrdquo state law should be rapidly developed to facilitate na-tionwide adoption of an appropriate law and uniformity of legal requirements
Summary of Analysis
Supporting Effective and Informed Adoption
The COVID-19 pandemic and the physical distancing efforts imple-mented to slow the rate of transmission have caused severe harm to indi-viduals communities and our society To protect the public good going forward we need a robust public health response that reduces the spread of SARS-CoV-2 and does so in a way that allows economic recovery to occur and to be sustained We also need to design and manage this public health response so as to minimize harms to individuals and society to distribute benefits and burdens equitably across the population and to avoid misuses of the technologies and the data they collect
To reduce the spread of SARS-CoV-2 chains of transmission need to be broken To do this people who have been exposed to SARS-CoV-2 or potentially exposed need to be identified as comprehensively and as quickly as possible so they can quarantine themselves and avoid infecting others This is the job of manual contact tracing by public health authori-ties in which people infected or presumptively infected with SARS-CoV-2 are interviewed and asked about their movements and interactions in-cluding where they work and shop how they travel with whom theyrsquove had contact and the nature of that contact (eg where the contact took place) Their contacts are then interviewed and potentially asked to quar-antine seek testing and take other protective measures if the contact is sufficiently high risk
14 Digital Contact Tracing for Pandemic Response
The hope is that DCTT can augment traditional contact tracing ef-forts either by working alongside and independently of manual contact tracing or by being integrated into manual contact tracing efforts in a way that makes these efforts faster more thorough and more efficient
Data suggest that a substantial proportion of transmissionsmdashper-haps as high as 50mdashoccur between individuals who are not symptom-atic and that transmission may occur as early as 3 days before onset of symptoms (WHO 2020) Because asymptomatic spread of SARS-CoV-2 appears to be a significant source of infection we need to identify po-tentially infected people before they show symptoms thus speed is of the essence This is one benefit of using DCTT potential contacts can be identified instantaneously notified quickly and asked to quarantine as soon as possible
Another benefit is identifying contacts who manual contact tracing methods may miss either because COV+ people do not remember all the places theyrsquove been or cannot identify all the people theyrsquove had contact with This is especially relevant given the long period of infectivity of SARS-CoV-2 which begins before people are symptomatic and aware they are infected (Ferretti et al 2020) If DCTT were designed to have optional location-monitoring capabilities this critical challenge could be mitigated even further For example location data might reveal that a COV+ person was at a restaurant at an exact time and date which could be followed up by contact tracers who could alert the public or use other measures to reach those who were also present in the restaurant at the same time In other disease contexts geolocation data have demonstrated some potential to support epidemiology and disease surveillance (see Fur-lanello et al 2002 Dredze et al 2013 Eckhoff and Tatem 2015 Fraccaro et al 2019) with technical cautions regarding accuracy and the like (Beu-kenhorst et al 2017)
One role for DCTT is to work alongside manual contact tracing but independently of it Individuals would download proximity tracing or exposure notification apps use them receive alerts if theyrsquove had a poten-tial contact with another user who is COV+ or presumptively COV+ and voluntarily self-quarantine without having contact with public health au-thorities or giving them data that feeds into public health contact tracing efforts It is possible that this would help to break chains of transmission and reduce the spread of SARS-CoV-2 though at this point these benefits
Summary 15
are speculative It is also possible that such exposure notifications will result in high rates of false positives
Another possible role is for DCTT to be integrated into manual con-tact tracing efforts When potential contacts are identified by DCTT they are connected to public health authorities who can then follow up with them There are different forms this could take and different kinds and amounts of data about contacts public health authorities could receive from DCTT On one end of the spectrum of reporting public health authorities would not receive individualsrsquo names or contact information only anonymous data The fullest version of reporting would securely send to public health authorities the names contact information such as address and phone number and other data about contacts that DCTT collected including data about their location and movement history
It is uncertain whether providing public health authorities with vol-umes of information on cases and contacts from DCTT will be useful in practice As mentioned above providing public health authorities with location data on cases and contacts collected by DCTT may help con-tact tracers to find and notify additional contacts However at present providing public health authorities with large amounts of data will be useful only if there is sufficient capacity to follow up on these data In addition there is a risk of low-quality data from DCTT flooding the system leading to investigation of false case contacts identified by DCTT and distracting from other important efforts Whether and to what extent data from DCTT will benefit contact tracing efforts is unknown pointing again to the importance of continuously collecting high-quality evidence about DCTT
Designing Flexible Technology to Maximize
Public Health Utility While Respecting Other Values
Use of DCTT is essentially an experiment as we have insufficient infor-mation about the performance of different DCTT and their efficacy In the face of this uncertainty how should DCTT be designed and how should its use be managed
Many efforts to advance DCTT in the United States and elsewhere have emphasized the importance of ldquoprivacy by designrdquo that is building privacy and security protections into the design of technology rather than counting on responsible use alone (Cavoukian 2010) As noted above
16 Digital Contact Tracing for Pandemic Response
some major technology companies have signaled this position through de-velopment of PPPT systems that embed features such as decentralization de-identified information user anonymity bans on collection of location data and minimal reliance on or integration of public health authorities or other government actors Many of these features have also been em-braced early by advocacy organizations (Crocker Opsahl and Cyphers 2020 Electronic Privacy Information Center 2020 Kahn Gilmor 2020) and in an open letter (ldquoJoint Statement on Contact Tracingrdquo 2020) from nearly 300 researchers These same actors have emphasized that use of DCTT should be fully voluntary
Although privacy is a key value individuals and communities may also value efficiency equity liberty autonomy economic well-being com-panionship patriotism or solidarity among other values People may accept more significant encroachments on privacy now if this ultimately results in realizing other values (such as companionship) that are of equal or greater importance to those individuals Rather than centering pri-vacy alone in design a different orientation is needed at this moment that of ldquovalues in designrdquo which incorporates a broader range of values into technology (Flanagan Howe and Nissenbaum 2008 Knobel and Bowker 2011) For example some users might wish to express autonomy solidarity or patriotism through DCTT by sharing their location history with public health professionals in order to advance the public health re-sponse increase system efficiencies (eg by contributing information that can lead to better data processing) and reduce the burden on essential workers At the same time there is value in further advancing autonomy by designing technology to allow individuals some control over what data about them are collected and shared
DCTT should be designed to have a base set of features that protect privacy and strive for interoperability but also should include other op-tional capabilities This could be achieved by designing DCTT to have a default that can be modified for example an initial setting could be that usersrsquo location data are not shared with public health authorities but us-ers may opt-in to this feature Such an opt-in approach is likely consistent with existing federal privacy laws
Designing DCTT this way gives users the flexibility to decide how to use the technology and how to engage with public health authori-ties consistent with their values and trade-offs they are willing to make This flexibility could also allow for more real-world evaluation of how
Summary 17
different users experience different features of DCTT in different loca-tions Technology design should not be static but it should be capable of evolving depending upon local conditions new evidence and changing preferences and priorities
DCTT developers must comply with a number of federal privacy laws These privacy laws generally permit the collection storage and use of personal information so long as the user provides meaningful consent Privacy law in the United States is generally sector-specific and limited in scope resulting in a patchwork of protections that differ significantly depending on the entity that collects the data and the type of data col-lected Given the complexity of existing federal privacy law and the need to further strengthen public trust in DCTT it would be beneficial for Congress to enact new privacy legislation that is specifically tailored to the use of DCTT in response to COVID-19 Such COVID-specific legisla-tion should be sensitive to the full range of values and recommendations described above
In short designing ldquomiddle-groundrdquo DCTT for flexible use may pro-vide the most adaptable and thus most robust public health responsemdashrespecting privacy and individual autonomy by allowing users to use DCTT in ways that express their own values
Public Acceptance of DCTT
While some groups have maintained that only PPPT-like minimal systems will be widely adopted because only they will earn and maintain public trust (Simpson and Conner 2020) there is insufficient evidence that pub-lic trust would be threatened by a DCTT system that has the capacity to securely collect location data integrate public health authorities and en-able voluntary sharing of certain user data (eg location data) with those authorities More research including through deliberative engagement sessions is needed to better understand how differences in the features and functionality of DCTT (such as optional sharing of geolocation data) influence trust and peoplersquos willingness to use DCTT Technology com-panies should not alone control the terms conditions and capabilities of DCTT nor should they presume to know what is acceptable to members of the public
Significant concerns have also been expressed by privacy advocates (Guariglia 2020) and in the popular press (Giglio 2020) about ldquosurveil-lance creeprdquomdashthat is a belief that state or corporate actors will use new
18 Digital Contact Tracing for Pandemic Response
surveillance technologies capacities and permissions well beyond the purposes for which they were initially justified to the public and beyond the time when they are useful for the COVID-19 pandemic Surveillance creep is a serious concern and should be carefully guarded against how-ever the possibility of surveillance creep is not a sufficient reason to limit development of DCTT to minimal systems Instead protections should be put in place to ensure that only those data that are necessary and relevant for the public health purposes at hand are collected and used and data should be kept only for the period of time needed for those pub-lic health purposes For this reason we would support COVID-specific legislation that would impose strict limits on the use of DCTT data for nonndashpublic health purposes
Finally the use of DCTT during the current pandemic should not set a precedent for future public health use (eg use in seasonal flu surveil-lance efforts) Future use would require independent justification Further use of DCTT in other contexts (eg by law enforcement or immigration enforcement) is presumptively unethical
Encouraging Adoption of DCTT
Researchers have estimated perhaps conservatively that DCTT use by 80 of smartphone owners or 56 of the population overall will be needed to suppress the epidemic (Hinch et al 2020) These estimates also highlight that some decrease in transmission would be realizable even with lower rates of technology adoption
In the United States many advocates and researchers have argued that use of DCTT must be fully voluntary However experience from other countries suggests that when use of a digital contact tracing app is voluntary only a minority of the population will download it Instead of making use fully voluntary and initiated by users there are ways that DCTT could be put into use without usersrsquo voluntary choice For exam-ple use of an app could be mandated as a precondition for returning to work or school or even further to control entry into a facility or trans-portation (such as airplanes) through scanning of a QR code to demon-strate personal exposure levels (Gan and Culver 2020)
While these approaches are hard to imagine in the United States some have argued that mandatory use of DCTT could be ethical If man-dates increase adoption of DCTT and improve the public health response
Summary 19
this would reduce the likelihood of lockdowns which are harmful and a severe limitation of individual liberty applied on a mass scale On the other hand mandated use of DCTT systems may not be effective People may not adhere to the mandate by simply leaving their phone at home Perhaps more important should the technology not deliver the hoped-for benefits having mandated the use of an unproven technology could result in a loss of public trust in the technology in the entity instituting the mandate and in the larger public health response potentially lead-ing to noncompliance with public health recommendations more broadly (Bernstein et al 2019)
Any decision maker considering mandatory use including govern-ment officials institutional leaders and employers must convincingly address a number of considerations Particularly important is the need to identify reliable evidence that the DCTT would be effective and to ensure that the burdens and benefits of use are equitable and justifiable At this time mandated use of DCTT by states or institutions is not jus-tifiable given uncertainty about potential harms and benefits Additional technology user and real-world testing is needed before mandatory use should be considered
As with any public health effort the amount of evidence that must be offered to illustrate that the intervention or program can achieve its aims and the degree to which people should be able to exercise choice in their participation should be in proportion to the anticipated bur-dens of the intervention or program For example the permissibility of mandating use of DCTT by the public depends on factors such as the sensitivity of the data that are collected the extent to which public health is integrated within the DCTT system and what actions are taken in response to confirmed virus exposure or being identified as COV+ (eg forced quarantine) The more burdens that are placed on individualsmdashfor example whether people are ordered into quarantine if they have been exposed to the virus or if there are limited social supports for those in quarantinemdashthe greater the demand should be on the performance of the DCTT system
Perhaps the most effective way to generate widespread US adoption of DCTT will be to offer incentives for its use in other contexts gener-ally speaking small incentives have been shown to lead to an increase in desired outcome (Singer and Ye 2013 Lee et al 2014) Given the impor-
20 Digital Contact Tracing for Pandemic Response
tance of widespread use modest incentives ought to be considered if and when there is sufficient evidence of the utility of DCTT so long as those incentives are not mandates in disguise Another ldquofirst linerdquo approach to increasing use of DCTT is for trusted community leaders public figures health care professionals and other respected individuals to communicate with the public and their communities about DCTT and to encourage its use through public engagement campaigns if and when the technology demonstrates sufficient potential
Establishing Meaningful Processes for
User Disclosure and Authorization (Consent)
Any effort to roll out DCTT should ensure that users have a meaning-ful opportunity to review and understand information about the specific technology and its uses Moreover given the importance of public trust and the current crisis of public trust in governments and technology com-panies handling private digital information there is a strong ethics argu-ment for requiring consent from individual users We recommend a care-fully crafted version of what is sometimes called simple consent which consists of basic disclosure and voluntary agreement or authorization This disclosure should include information about the purposes of the technology the userrsquos options for collecting and sharing data purposes for which data can be used and any known risks among other informa-tion This information should be presented in an accessible format on any DCTT app and more detailed disclosures should be readily accessible for those who wish to review them
Through an opt-in mechanism such as clicking a button to signal agreement users should be able to indicate their intention to use a DCTT The opt-in approach is consistent with mechanisms for agreement to use other downloaded applications An opt-in approach should be part of the initial introduction of DCTT given the novelty of the technology and its uses and the need to build trust and confidence in the system Successes of opt-out approaches in other areas suggest that the feasibility and value of an opt-out approach to DCTT should be carefully evaluated particularly in conjunction with assessment of whether public health goals are being met (Rithalia et al 2009) Such assessments should be informed by what is technologically possible by local data regarding benefits and harms of the technology and by evolving understanding of the degree to which
Summary 21
an opt-out approach is likely to increase or decrease utilization among different populations
Promoting Equity and Fairness in Application of DCTT
Digital contact tracing technology should be designed and used in ways that as far as possible promote an equitable distribution of benefits and burdens DCTT should be deployed in a manner that does not propagate preexisting patterns of unfair disadvantage or distribute harms and risks unfairly throughout the population It is well known that some commu-nities have lower rates of technology and data access and therefore may benefit less from use of DCTT unless steps are taken to address these digital disparities Additionally should use of DCTT be made a require-ment for entry into a workplace into a school or onto transportation then those who currently do not possess the required technology must not be unfairly burdened through lack of access In order to mitigate this states localities and institutions that recommend widespread use of DCTT should provide technology (eg mobile phones Bluetooth de-vices) and free data packages to members of the community who desire but lack access to these devices
Some populations may also experience greater harm and greater fear of harm from having their data collected For example some groups such as African Americans Hispanic Americans Muslim Americans and undocumented immigrants have more reasonable fear of their data being handed over to law or immigration enforcement and some groups have lower levels of trust in public health due to past injustices (Auxier et al 2019 CSM 2017 Pew Research Center 2017 Rodrigues et al 2018) This further substantiates the need to limit use of any data gathered by DCTT to its public health purpose
Instituting Transparent Governance and Oversight
DCTT must be developed with an eye toward both present and future implications We are rapidly gaining knowledge about SARS-CoV-2 and COVID-19 but still have essential gaps in our understanding In the United States public health responses including DCTT will generally be developed and coordinated by individual states regional consortia (Res-ton Sgueglia and Mossburg 2020) and associations Good governance in this context requires transparency and the creation of oversight bodies
22 Digital Contact Tracing for Pandemic Response
with the appropriate expertise and representation to allow nimble and effective responses while serving as trusted representatives
In order to address the range of ethics and governance concerns that relate to the design and use of DCTT we recommend that digital surveil-lance oversight committees be established perhaps at a state level and with a platform for national coordination These committees can provide ethics and regulatory review prior to and concurrent with widespread use of DCTT The committees should be composed of a diverse group of experts capable of evaluating the quality of a DCTT system locally including members of communities that experience higher rates of digital disparity
When assessing the design and use of digital contact tracing systems these committees (and the public more widely) should consider not only the risks and benefits accrued during the COVID-19 pandemic but also implications for the future How can we navigate safe use of these tech-nologies in a way that preserves public trust in them and enables the possibility of future beneficial use
As a start it should be emphasized that the principles offered in this and other guidance documents do not apply only during the pandemic Future efforts to advance DCTT capabilities during quieter times should make every effort to follow them
23
Introduction
Public health professionals around the world are working tirelessly to respond to the COVID-19 pandemic using tried-and-tested public health methods for infectious disease surveillance and control These traditional methods are essential to the global COVID-19 response To complement these actions and potentially augment the speed and efficacy of the public health workforce digital technologies are being harnessed Given the scale of the pandemic significant efforts are being undertaken to develop and leverage public-facing and health-system-supportive tech-nology solutions including smartphone apps and other digital tools that may aid public health surveillance and contact tracing
Digital contact tracing technology and closely related digital health products (hereafter DCTT) have been used in several countries as part of broader disease surveillance and containment strategies Globally many digital COVID-19 contact tracing strategies have already emerged in response to the pandemic This is not surprising given the ubiquity of mobile phones and other digital devices around the world (ldquoMobile Cellular Subscriptions (per 100 People)rdquo 2018) experiences developed during prior outbreaks and pandemics and the pre-COVID-19 momen-tum behind using digital technologies to support individual and health system capabilities (WHO 2017 Mathews et al 2019 Aiello Renson and Zivich 2020 Mahmood et al 2020) In the United States DCTT has been proposed as an integral part of some plans to ldquoreopenrdquo the country (Allen et al 2020 Hart et al 2020 Simpson and Conner 2020) It is al-most certain that these and related technologies will become part of not only the COVID-19 response but also the larger toolbox for future public health communicable disease prevention and control
24 Digital Contact Tracing for Pandemic Response
While novel public health surveillance technologies such as DCTT have theoretical promise their effectiveness is unclear These technologies also raise important ethical legal and governance challenges that require comprehensive analysis in order to support decision-making regarding their appropriate use A number of frameworks recommendations and analyses have emerged recently in an effort to chart potentially ldquosaferdquo pathways for use of public health disease surveillance technology Many in the United States such as the Electronic Frontier Foundation Elec-tronic Privacy Information Center American Civil Liberties Union and the Center for American Progress are proposing that digital public health surveillance technologies must embrace strict data privacy protections decentralized data storage a high degree of anonymity and voluntary adoption (Crocker Opsahl and Cyphers 2020 Electronic Privacy Infor-mation Center 2020 Kahn Gilmor 2020 Simpson and Conner 2020) Others have argued that technologies that seek to enhance public health response during a pandemic should more closely align with the needs of public health professionals and the evidence-based procedures they follow stating that interests in serving the publicrsquos health ought to weigh more heavily in the necessary balancing of stakeholder interests (de Jong et al 2019 Watson et al 2020) This view is in part based on a recogni-tion that during countless other outbreaks the public has benefited from traditional disease surveillance and contact tracing which are heavily re-liant on centralized data storage and when necessary the collection of identifiable information These traditional approaches are governed by ethics principles (PHLS 2002) ethics guidelines (WHO 2017) and laws (ASTHO 2012) and digital technologies represent a new tool to support them
While debates and recommendations about appropriate design and use of DCTT have focused intensely on minimizing important data-related risks a wider lens is needed to fully appreciate the many additional criti-cal questions that need attention This report begins to grapple with these questions which are critical to address in order to guide responsible use of DCTT Given the complexity of the terrain as a first step toward estab-lishing a foundation for responsible decision-making regarding potential use of DCTT we offer a set of guiding principles (see box) These prin-ciples are meant to apply to DCTT as well as other digital technologies used in novel ways during pandemic response
Guiding Principles for the Use of Digital Public Health Technologies for Pandemic Response
Transparency and public engagement are essential to an inclusive digital public health response
bull Government public health and digital technology leaders must engage
effectively with the public and other stakeholders to communicate the util-
ity importance oversight and limitations of relevant digital technologies
including their implications for individualsrsquo privacy and civil liberties
bull Transparency at all levels is essential for maintaining public trust and
confidence
bull To the extent possible digital public health responses should reflect the
range of values that are important to individuals including advancing the
health and well-being of the community as a whole
bull Decision makers should recognize the sacrifices that some people may be
willing to make during a pandemic in order to advance public health goals
Acceptance by some of particular monitoring capabilities should not be read
as a willingness to extend these methods to other problems or uses
Digital public health responses must represent the least infringement of civil liberties necessary to accomplish the public health goals
bull If preferred digital public health strategies infringe on privacy and other civil
liberties the infringements must be sufficiently justified by the circum-
stances of the pandemic offset by ample anticipated public benefit and
considered relative to infringements associated with other possible strate-
gies such as mass physical distancing
bull Only those data that are necessary and relevant for the stated public health
purposes should be collected Identifiable data should be stored in a secure
manner and only for the period of time that the public health purposes
require
bull Adopted technologies should not be used in ways that subject communities
to discrimination or surveillance for nonndashpublic health reasons
bull Respect for individual autonomy requires that users are sufficiently informed
of the public health goals of the technology and the extent to which those
goals are being met
Use of digital public health technologies and data must be guided by best available evidence
bull Decisions to deploy digital public health technologies should be based on a
careful assessment of the uses and limitations of any proposed technology
taking into account the best available evidence
bull Those who deploy digital public health technologies should continuously and
systematically monitor their performance as well as any evidence that is being
generated in other contexts about the selected technological solution and about
other competing technologies
bull Unintended consequencesmdashincluding those that might impact public health
goals core values and interests of the public and unfair advantage or disadvan-
tagemdashshould be carefully monitored and addressed as necessary
Responsible use of digital public health technology requires meaningful governance and accountability
bull Systems of governance must be trustworthy and well informed They must be
reviewed and adjusted as circumstances and evidence change or as unintended
effects are identified
bull Trusted representatives who are capable of developing and implementing uni-
form and fair standards for adopting and utilizing underlying digital technology
must be identified
bull Understandable transparent and publicly accessible rules must guide the
collection access control use storage and combination of data by government
authorities public and private institutions and other parties such as public
health researchers
bull Oversight accountability and consequences for abuse or misuse of these data
must be explicit and enforceable
The deployment of digital public health technology must be rooted in a commitment to equity
bull Digital public health technologies should be deployed in a manner that does
not propagate preexisting patterns of unfair disadvantage or further distribute
harms and risks unfairly throughout the population
bull To the extent possible digital public health technologies should be designed to
rectify existing inequities
bull Oversight mechanisms must be in place to ensure that the improved public
health outcomes are equitable and to detect and correct any unforeseen resul-
tant injustices attributable to the technology or that can be addressed using the
technology
bull The incentives and disincentives for adopting new technology must be equita-
ble not exploitative and aligned with effective use of the technology
bull Disparity-driven technology gaps should be explicitly recognized To the extent
possible provisions should be made to address the digital divide
Introduction 27
In reflecting on these principles it becomes clear that if we wish not only to realize but to maximize the public good that might come from use of DCTT we must carefully define and responsibly align public health needs and capabilities with technological needs and capabilities We must understand that although technology may serve as a workforce multiplier it alone will not solve the public health challenges we face We must identify and address assumptions and misinformation about technologies and data use We must provide the means and opportu-nity for informed decision-making by the public and those who serve as our representatives Government officials public health leaders leaders of other institutions employers digital technology developers and the public all must be adequately informed and engaged in order to make the best decisions possible under the circumstances
ONE
29
Types of Information Collected through Contact Tracing
Data Collected from Infected Persons
Symptoms and Course of Illness
Information about COVID-19 patientsrsquo signs symptoms and course of illness is important to public health because it provides a basis for refining clinical case definitions and informing health care providers and the gen-eral public (CDC 2020c) This includes the specific signs and symptoms manifested by persons who are COV+ as well as the relative frequency and durations of different signs and symptoms This would also take into consideration those persons with no symptoms but who test positivemdashthose who are presymptomatic (develop symptoms later) those who are postsymptomatic (clinically recovered but still infectious) and those who never manifest illness at all
Typically contact tracing begins with a case in which a person has confirmation of infection by means of a diagnostic test However in some cases test results are not reported until several days later and individuals may be identified as ldquopresumptive positiverdquo cases until testing can be completed In these cases contact tracing efforts will need to be updated when test results are returned For example if a test comes back negative public health professionals will want to notify contacts that they no lon-ger need to quarantine
Public Health Perspective
30 Digital Contact Tracing for Pandemic Response
Movement and Contacts
In order to manage cases appropriately (identify and track the infected isolate the sick quarantine the exposed) public health officials need de-tails on each case (Resolve to Save Lives nd) First they need to know who and where the individual is That means personally identifiable infor-mation and contact information (address phone numbers email) It also means information about the nature intensity and duration of contact with individuals to whom they may have transmitted the disease This may include information about where the individual works and the kind of work they do (eg health care worker) how they travel (eg bus sub-way car) and where they shop or any other public venues they may have visited during a period of possible infectiousness (PIH 2020a) It may be helpful in certain circumstances for public health officials to ensure that suspected cases contacts or other high-risk individuals are following iso-lation and quarantine recommendations or orders
Contact tracing involves identifying all individuals who have had sig-nificant exposure to confirmed or probable cases during the time prior to and after the onset of symptoms both of which are times when the case is thought to be infectious (Africa CDC 2020) Contacts could be those who are caring for COVID-19 patients especially if they lacked proper PPE and those who had close interaction with the COV+ person over a sustained period of time particularly in enclosed spaces (PIH 2020a) For COVID-19 contacts are identified by asking a person with a confirmed or probable case about people they may have been within 6 feet of for 15 minutes or more starting from 48 hours before the onset of symptoms and lasting until the person is isolated (CDC 2020b)
Data Collected from Contacts of Infected Persons
Contact Details
In addition to the data collected from individuals with COVID-19 con-tact tracers will collect data from potentially exposed individuals (con-tacts) Information about the nature intensity and duration of contact with an infected person may be collected for a contact if information about the case is known to the contact These details can help a contact tracer more accurately determine whether the contact is at high or low risk for SARS-CoV-2 transmission and help determine whether a con-tact should quarantine for 14 days (the upper bound of the SARS-CoV-2
Public Health Perspective 31
incubation period) In addition public health professionals may gather contactsrsquo demographic information and other personal data to contribute to population-level disease surveillance and situational awareness about an epidemic (CDC 2005) However the information needed at baseline is only a personrsquos name and contact information
Symptoms (If They Develop) and Course of Illness as well as Information about Close Contacts
If a contact develops COVID-19 symptoms while in quarantine andor tests positive for the virus public health will then collect the data required for a COVID-19 case This includes collecting information on the con-tacts that a person may have had (if any) in the days immediately before and during the course of their infection
How Contact Tracing Information Informs Public Health Action
To reduce disease burden and help make ldquoreopeningrdquo safer during the COVID-19 pandemic the United States and other countries will need to identify gather information about and safely isolate cases and quar-antine their contacts to reduce community transmission (Watson et al 2020) Gathering information about possible cases and their contacts en-ables public health to break chains of transmission
Contact tracing involves stages (CDC 2020a) including
1 identifying an infected person as a COVID-19 case
2 identifying the close contacts of that case (Africa CDC 2020)
3 getting in touch with contacts
4 asking contacts to quarantine at home for 14 days
5 assessing contacts for possible symptoms and
6 following up with COV+ persons and their contacts to identify new or worsening symptoms and connect them with medical care if needed
Contact tracers also play an important role in providing resources for COV+ persons who are in home isolation and their contacts who are
32 Digital Contact Tracing for Pandemic Response
in home quarantine Knowing who and where cases and contacts are can enable provision of supplies such as digital thermometers or masks Effective contact tracing that enables isolated cases and quarantined con-tacts to remain at home also requires providing a range of social sup-port services or ldquocare packagesrdquo from delivering food and medicines to trash pickup Furthermore vulnerable individuals who are homeless or otherwise unable to sufficiently isolate or quarantine in their current living conditions may need to have alternative housing arranged to safely remain separated from others (CDC 2020b)
Finally contact tracers explain what quarantined contacts should do if they begin to develop symptoms consistent with COVID-19 (Africa CDC 2020) Depending on the context contact tracers may engage in active monitoring by regularly communicating with contacts about their health status through phone text message or possibly mobile applica-tions In rare cases public health can make quarantine mandatory and may monitor a quarantined individual to ensure that they do not break quarantine Contact tracers may also facilitate access to health care by providing telemedicine resources or other information and support for accessing medical care
Characteristics That Make Data Useful to Public Health for Reducing Disease Transmission
Data Access
If digital contact tracing technology and closely related digital health products (together DCTT) are intended to support the public health ac-tions described above and directly amplify public health capacity to con-duct case identification and contact tracing then data collected through DCTT must be accessible to public health authorities Identifying infor-mation and location data for cases and contacts of cases are necessary for public health use so that contact tracers can do their work to uncover ongoing transmission and enable isolation and quarantine These data should also be durable meaning that public health can return to the data in order to interact with and support cases and contacts These data can also be useful at a population level if de-identified and aggregated by illuminating trends in community transmission and providing support for decisions about resource allocation
Public Health Perspective 33
Data Format
Data should be provided to public health authorities in a usable format that is compatible with public health systems and that has the granularity and specificity of personal information that is needed for use in contact tracing Without personal identifiers the data cannot be used by public health workers to undertake contact tracing Data should also contain information about the nature of a contact including the proximity of the contact and number of minutes that the person was in contact with an infected individual Location data can also help public health author-ities to conduct contact tracing particularly when contact occurred in a crowded area and involved people who donrsquot know one another Loca-tion data from a case can help public health professionals identify con-tacts even when those contacts themselves are not using a contact tracing app because the data shows contact tracers where to look for additional contacts
Data Accuracy
Data that identifies individuals as having sustained contact with a case must be as accurate as possible If criteria for being considered a contact are too restrictive it may result in missed contacts and sustained chains of disease transmission If criteria are too broad it may result in unneces-sary restriction of movement which could have significant personal and economic consequences
Timeliness of Data
Data from cases and contacts must be timely in order to enable case-based management that will help reduce community transmission For contact tracing to be effective infected individuals need to be isolated and their contacts identified and quarantined as quickly as possible Testing for SARS-CoV-2 can take time sometimes many days for a test result Especially because SARS-CoV-2 is transmissible during the pre-symptomatic period data on symptomatic individuals should be made available to public health officials even before a positive test is returned in order to enable identification and quarantine of contacts right away If this information is delayed until a test result is received it may be too late to identify and quarantine contacts because contacts (if infected) will already be contagious and may have spread the virus to others
34 Digital Contact Tracing for Pandemic Response
VolumeAvailability of Data
The more that individuals opt to share their information to support con-tact tracing the more effective contact tracing will be in breaking chains of viral transmission and controlling epidemics of COVID-19 The exact proportion of cases and contacts that need to be identified in order to avoid large surges of cases which overwhelm health care systems is un-certain but the goal is to identify all infected cases and all close contacts of each case (PIH 2020b)
Recommendations
bull Technologies or apps with the goal of enhancing public health ca-pacity to identify cases and trace contacts in order to control the spread of SARS-CoV-2 should be designed to match functionality with that goal
bull Technologies or apps may produce some false negatives or false positives but they should be accurate enough that public health authorities feel confident that they support and donrsquot detract from contact tracing efforts
bull DCTT approaches for public health should be designed to facilitate the following
deg identifying contacts including those who may not be easily found otherwise
deg finding and notifying contacts rapidly before they develop symptoms if infected
deg analyzing the nature of contact to determine whether contact is high medium or low risk and to support decisions about whether a contact should quarantine and
deg following up with cases and contacts so that public health can provide resources to support isolation and quarantine
bull Data collected through DCTT should be made available to public health professionals and to researchers in de-identified form to sup-port population-level epidemiologic analysis
T WO
35
Characteristics of SARS-CoV-2 Relevant to Candidate Digital Solutions
The SARS-CoV-2 virus has some unique transmission characteristics and clinical manifestations that can help guide use of digital contact trac-ing solutions Individuals infected with this virus may or may not show symptoms or may show a range of different and sometimes nonspecific symptoms Estimates regarding the percentage of individuals who are in-fected but never develop symptoms is highly uncertain ranging from 5 to 50 (Heneghan Brassey and Jefferson 2020) Data suggest that a substantial proportion of transmissionsmdashperhaps as high as 50mdashoccur between individuals who are not symptomatic and that transmissibility may extend out as long as 3 days before the onset of symptoms (WHO 2020)
The complexity of asymptomatic and presymptomatic transmission makes it more difficult to identify all cases of COVID-19 It also means that manual contact tracing is less effective because people are unlikely to remember all of their contacts during the long period of infectivity (Ferretti et al 2020) however it does not negate the need for contact tracing Identifying symptomatic cases will still help greatly with slowing the spread because their contacts can be asked to quarantine to prevent them from spreading the virus if they are indeed infected This means that whether they become symptomatic or not contacts will be quarantined and the chain of transmission will be broken If contact tracing can be implemented on a large enough scale perhaps with support from DCTT
Digital Technology and Contact Tracing
36 Digital Contact Tracing for Pandemic Response
eventually the virus could be managed at much lower levels of community transmission and large epidemics of unrecognized spread will not occur
The transmissibility of the virus when a person has no symptoms further suggests that effective solutions may require multimodal inter-ventions combining contact tracing with frequent rapid and ubiquitous testing and continued social distancing to varying extents (Cheng et al 2020)
Because of presymptomatic spread contact tracing efforts and dig-ital solutions to augment those efforts should support identification of contacts a person had 2 days before their symptoms and at least 3 days after the resolution of those symptoms (if the person continued to have contacts through that time period) (CDC 2020d) Additionally public health messages delivered by these technologies should urge contacts to quarantine for the full 14-day incubation period
Previously Existing Contact Tracing Technologies
Prior to this pandemic health agencies in high- medium- and low-income countries had begun to develop and use digital tools to augment the man-agement of infectious diseases including sexually transmitted infections (HIV chlamydia gonorrhea) and high-consequence epidemics (Ebola) (Danquah et al 2019)) However these have been primarily used to facil-itate case interviews partner notification (in the case of STIs) and record keeping as opposed to fully digitizing or automating the contact tracing process
It has been recently suggested that digital contact tracing could con-tribute to the management of the ongoing COVID-19 pandemic and the experiences of containing SARS-CoV-2 in countries such as China Sin-gapore and South Korea provide noteworthy examples However un-dertaking this case-based intervention on the scale required to achieve pandemic control is a novelty in the history of public health Although technological development is proceeding rapidly several foundational is-sues have yet to be resolved including functionality connectivity to pub-lic health authorities and informatics systems usability by disease inter-vention specialists (DIS also referred to as contact tracers) and sufficient protection of personally identifiable information among others
Digital Technology and Contact Tracing 37
Introduction of Novel Digital Contact Tracing Technologies
Digital contact tracing technologies and platforms have recently been in-troduced and the CDC has published preliminary criteria for evaluating these tools (CDC 2020e) It can be helpful to consider three broad ap-proaches along the spectrum of potential methods of digital contact trac-ing a maximal approach (typified by the South Korean governmentrsquos cen-tralized data collection (M S Kim 2020)) a minimal approach (typified by the AppleGoogle decentralized privacy-protecting proximity tracking (Apple and Google nd)) and a diverse middle ground that aims to aug-ment manual contact tracing with the collection of digital data Perhaps the most promising approach in this middle ground involves allowing us-ers to turn over both proximity data and GPS location data (ie cell-site location data) to public health authorities on a voluntary basis
Along with this ldquominimal to maximalrdquo spectrum in the design of dig-ital contact tracing technologies and systems there is another spectrum that concerns voluntary versus mandatory use of these technologies are individuals entirely free to use these technologies or not or should poli-cies incentivize or even mandate their use At one extreme South Korea (Republic of Korea) implemented a system (called Safe Korea) supported by the Ministry of the Interior and Safety that collects a variety of per-sonal data in a centralized database in order to enforce quarantine orders and track possible contacts (M S Kim 2020) Israel also implemented a centralized involuntary data collection system for tracking COVID-19 cases and alerting those who may have been exposed (Hendrix and Eg-lash 2020) In Poland health authorities have set up mandatory ldquocheck-insrdquo involving a GPS-waypoint capture and ldquoselfierdquo photographs sent to the monitoring agency to ensure that individuals are not breaking quar-antine (Hamilton 2020)
These centralized systems can be designed to incorporate data from a variety of sources The data collected include location data from mobile phones QR codes can also be scanned to track the use of public transit where GPS data may be inadequate (due to low resolution) to accurately distinguish the occupants of one vehicle from another The data collected from mobile phones can then be integrated with data from other sources such as facial-recognition cameras credit card transactions and social media
38 Digital Contact Tracing for Pandemic Response
At the other extreme of technology invasiveness for contact trac-ing isolation and quarantine many corporations and working groups (including the AppleGoogle collaboration) have developed privacy-pre-serving proximity tracking (PPPT) using Bluetooth Low Energy (BLE) ldquohandshakesrdquo that record close contact between mobile phone users In most architectures these proximity data are stored in the usersrsquo phones as anonymized ldquobeaconsrdquo that cannot be used to re-identify the users directly If a user with a PPPT app installed on their phone tests positive and enters test results into their app those who have been identified as having been in close proximity to them can be notified by the app This notification can be automatic or at the discretion of the person who is COV+ depending on the app design If notified a user who has been in contact with a COV+ individual would receive a push notification alert-ing them to possible exposure (which may be timestamped) but with no other identifying information
Because of its reliance on anonymized data PPPT on its own is dis-tinct from manual contact tracing In recognition of this fact some de-signers and researchers now use the more descriptive term ldquoexposure no-tificationrdquo Moreover the public health usefulness of PPPT is uncertain it is unclear how PPPT can best be used in tandem with manual contact tracing especially if the data it collects are inaccessible to or unusable by public health authorities It remains to be seen whether PPPT will provide significant benefit operating alongside but not integrated into manual contact tracing
Between these extremes there are a number of possible middle-ground approaches that aim to strike a balance among public health utility tech-nological feasibility and user privacy protections This middle ground divides into two rough categories centralized storage of de-identified data and decentralized storage of personally identifying data The United Kingdomrsquos NHSX is reportedly developing an app that would utilize BLE handshakes to collect anonymized proximity data which would then be stored on a centralized government-operated server
The most prevalent middle-ground approach in the United States context involves the collection and storage of personal datamdashincluding identifying information and location datamdashon the userrsquos phone This de-centralized but personally identifiable data can then be voluntarily shared with public health officials if the user tests positive for SARS-CoV-2 For
Digital Technology and Contact Tracing 39
example an MIT team has developed an app called Private Kit Safe Paths (MIT nd) that stores usersrsquo location data on their phone for 28 days If a user tests positive she can voluntarily upload her location data to a website that is accessible only to public health officials Officials can then analyze this personally identifiable data and subsequently broadcast re-dacted and de-identified data to other users Healthy users would have access to these redacted location data of COV+ users but their own data would not leave their phones (The developers plan to incorporate BLE proximity data once available) Along similar lines the North Dakota state government has rolled out an app that stores both location data and proximity data on a userrsquos phone which can be voluntarily released by the user to public health authorities if the user tests positive (NDDoH 2020) At a minimum the storage of user location data can function as a ldquomemory aidrdquo if the user tests positive But releasing the data to public health authorities may help them analyze the spread of COVID-19 and alert individuals or groups that have been in contact with persons who are COV+ An overview of various DCTT apps and platforms as well as features that are relevant to this analysis are provided in Table 1
Because DCTTs are so new very little is known about their actual utility to public health authorities for controlling this pandemic Although multiple countries that have had success in greatly reducing transmission of SARS-CoV-2 have included DCTT in their response these countries have employed multiple simultaneous approaches to controlling the vi-rus including manual contact tracing and it is difficult to disentangle what made those responses successful Preliminary impressions from Ice-land may suggest that DCTT at least in that context had a small impact on reducing transmission ldquoespecially compared with methods of manual contact tracing such as phone callsrdquo (Hadavas 2020) This is with the highest public download rate of any DCTT app thus far
DCTTs have the potential to be helpful but they also have the poten-tial to distract from other public health efforts including manual contact tracing Concerns about implementation of DCTT from the public health perspective include that data generated may not be useful to public health authorities either because they donrsquot include detailed data to aid contact tracers or because the data are inaccurate (Mills Rodrigo 2020) DCTT if not calibrated well could be overly inclusive and create many false positives This would be harmful to those individuals being notified and
Purp
ose
Tech
nolo
gies
Use
dD
ata
Stor
age
Part
icip
atio
n
Proximity-based exposure notification
Digital contact tracing (DCT)
Bluetooth LE
GPS
SMS
Centralized
Decentralized
Mandator (actually or functionally)
Voluntaryopt-in
Gov
ernm
ent a
cces
sA
pp N
ame
Dev
elop
er o
r Cou
ntry
Max
WeC
hat
Alip
ayCh
ina
D
ata
com
es fr
om g
over
nmen
t so
urce
s lo
catio
n da
ta s
ent t
o po
lice
Intervention Type
Middle Ground
Trac
e To
geth
erSi
ngap
ore
M
anda
tory
gov
ernm
ent
acce
ss if
pos
itive
NH
SXO
xfor
dO
xfor
d
Gov
ernm
ent m
aint
ains
dat
a
Nex
tTra
ceFr
ed H
utch
inso
n Ca
ncer
Re
sear
ch C
ente
r
Gov
ernm
ent m
aint
ains
dat
a
but n
o st
orag
e
COVI
D
Safe
Path
sM
IT
Volu
ntar
y up
load
by
user
s
who
test
pos
itive
Aar
ogya
Set
uIn
dia
A
nony
miz
ed a
ggre
gate
Care
19N
orth
Dak
ota
In
agg
rega
te o
ptio
nal i
f pos
itive
Minimal
Covi
dSaf
eU
niv
of W
ashi
ngto
n
Non
e
Covi
dWat
chU
niv
of S
tanf
ord
amp U
niv
of W
ater
loo
To
val
idat
e te
st re
sults
CoEp
iCo
Epi
O
pt-in
to s
hare
BT
and
sym
ptom
lo
g w
ith C
oEpi
ser
ver
itoG
erm
any
N
one
pos
itive
resu
lts to
ito
serv
er
TAB
LE 1
Ex
ampl
es o
f Dig
ital C
onta
ct T
raci
ng T
echn
olog
ies
to S
uppo
rt A
ctiv
e Pu
blic
Hea
lth S
urve
illan
ce a
nd R
elev
ant F
eatu
res
Digital Technology and Contact Tracing 41
asked to quarantine unnecessarily and it could result in large proportions of the population remaining at home at any one time Individuals living or working in congregate settings could receive frequent notifications that would result in their inability to leave quarantine for long periods of time Finally public health authorities could also become inundated by data from these technologies and not have sufficient approaches to manage or analyze the incoming information
Relevant Differences between Manual and Digital Contact Tracing
There are several noteworthy differences between manual contact tracing efforts and use of DCTT First there is a significant amount of evidence regarding the effectiveness of manual contact tracing which is lacking for DCTT Second manual contact tracers interact with individuals who are confirmed or suspected cases and contacts of cases but not other members of the general public DCTT intervention would affect all users regardless of circumstances (though some more than others) Third manual contact tracing occurs most often through human-to-human encounters with the opportunity to clarify misconceptions address worries and express sympathy and other important affects DCTT can certainly incorporate sharing of important information and potentially communicate some af-fect but it currently lacks a range of other human capabilities and char-acteristics Fourth there typically are fewer data intermediaries in manual contact tracing (fewer entities handling data) in DCTT a valid argument could be made that a wide range of technology developers (and perhaps mobile network operators) must remain connected to relevant data in order to continuously identify problems and improve functionality
It is because of these and other differences that DCTT has been pro-posed as a potential complement to rather than a replacement for man-ual contact tracing However over time it is possible that technology could develop to close gaps between some of these differences (if and as needed) and in parallel the goals of contact tracing and public health surveillance may evolve
Ethics of Designing and Using DCTT
43
THREE
43
Those developing DCTT and those considering its use should systemat-ically take into account and document alignment with the guiding princi-ples outlined in this report
When considering the ethics of DCTT key ethical questions con-cern the features that DCTT should have (eg should digital contact tracing apps collect usersrsquo location data) whether and how individualsrsquo data should be shared with public health authorities how ethically to encourage use of DCTT (eg under what circumstances would it be eth-ical to incentivize or mandate use of DCTT) what kind of supports and equity-promoting measures should accompany use of DCTT and how governance and oversight of DCTT should be structured
The sections that follow consider these questions one by one A key conclusion of this report is that these features of the design and use of DCTT are ethically interrelatedmdashreaching a determination regarding any one question requires careful consideration of them all Rather than reaching ldquoone size fits allrdquo conclusions about specific features of uses of DCTT decision makers should ethically assess DCTT systems holistically
Generally a public health measure is ethically justifiable if it strikes a reasonable balance between competing considerations and if it pro-vides sufficient public health benefit (or the prospect of benefit) to justify the burdens associated with it DCTT systems are ethically justifiable if they strike a reasonable balance between multiple ethical considerations including
bull enabling an effective and efficient public health response
bull protecting individual privacy and preventing harms to individuals
44 Digital Contact Tracing for Pandemic Response
including harms from sensitive data being revealed and from erro-neously being subjected to isolation or quarantine orders
bull allowing individuals to control what information about them is collected and revealed to whom including through appropriate dis-closure and authorization processes for data collection
bull promoting equitable distribution of benefits and burdens of DCTT
bull maintaining public trust in DCTT and in the COVID-19 public health response and
bull taking seriously the future implications of decisions that we make today
To illustrate a holistic assessment consider whether it is ethically jus-tifiable for an employer to mandate that employees use a DCTT as a condition of returning to work This will depend upon many features of the DCTT system what kind of data the DCTT collects (eg does it collect location data or just record proximity events) whether there is public health capacity to make good use of these data what the data are used for (eg will the employer ban an employee from the workplace on the basis of a DCTT-identified contact) what kind of social supports are available (eg is there paid leave for employees) what employeesrsquo attitudes are toward use of DCTT and whether mandating use is likely to have public health benefit among other factors These factors may vary from place to place and may change over the course of the pandemic Thus there is no ldquoone size fits allrdquo ethically optimal approach to DCTT
Justifying the Use of DCTT Systems
A foundational issue is why deploying any DCTT during a pandemic is justified given there are manual contact tracing capabilities that are well established while the performance and effectiveness of novel technolo-gies is less established The need to move quickly to minimize the spread of the virus poses challenges here as the data needed to fully make the case that these technologies substantially contribute to the public health response may not be available prior to widespread use The primary ar-gument for DCTT is that the capacity of manual contact tracing may be
Ethics of Designing and Using DCTT 45
exceeded and we may not be able to bolster the public health workforce rapidly and sufficiently enough to meet needs DCTT has the potential to quickly and exponentially expand the reach of contact tracing In ad-dition DCTT may allow more efficient identification and quarantine of potential contacts of COV+ people than manual contact tracing alone particularly given the high number of infections that have been spread by asymptomatic individuals
Nonetheless reasonable people disagree about the prudence of pur-suing DCTT especially given its limited performance history and poten-tial risks including diverting attention and resources from more effective interventions The limited attention and resources available during a pan-demic must be allocated efficiently and effectively
To justify potentially widespread use of technologies such as DCTT therefore a number of considerations must be addressed
bull whether the technology is designed to meet an important and unmet public health need
bull whether there is sufficient evidence or reason to suggest that the technology will be effective at serving its purpose
bull whether the outbreak is characterized by sufficiently severe morbid-ity and mortality and a high rate of disease transmission to warrant large-scale introduction of novel systems
bull whether there are other less autonomy-restricting or less risky al-ternatives to widespread use and
bull whether it is reasonably likely that a sufficient number of individ-uals will use the technology to achieve the intended public health benefit
Monitoring and Evaluating Technologies to Inform Policy and Practice
A number of public health ethics principles necessitate the ongoing mon-itoring and evaluation of DCTT systems First DCTT must be shown to perform reasonably well at achieving its stated goal reducing the spread of SARS-CoV-2 The effectiveness of DCTT programs should be illus-trated at a number of stages
46 Digital Contact Tracing for Pandemic Response
1 Robust initial technology testing is needed to publicly justify the widespread adoption of DCTT and avoid public failures which may hamper future uptake (eg Lovejoy 2020 Morse 2020) This typically includes alpha testing in virtual environments and beta testing in different community settings
2 If and when a DCTT is implemented on a wide scale it must be monitored on an ongoing basis to assess reach effectiveness func-tionality best practices and any harms
3 When approaching a previously identified stopping point for use of DCTT monitoring can help to identify when utilization is no longer needed
If at any of these points evidence clearly suggests harm (particularly in comparison to other methods that the public might find more acceptable) this evidence should provide a basis upon which to revisit strategies pri-orities and allocation of resources Attention should be given to foresee-able side effects that may dramatically influence the overall effectiveness of the program such as individuals carrying their smartphones around with them selectively so as to avoid particular undesired consequences of DCTT policies
Anonymized aggregate data including user feedback must be eval-uated to ensure that benefits and burdens are distributed fairly As noted earlier unintended burdens may include inequitable outcomes that may arise in a DCTT program for example resulting from uneven access to the required technology to participate disparate concerns about sur-veillance within some communities that might limit widespread use or discrimination that may result from being identified as COV+ due to the program or for communities that are termed ldquohotspotsrdquo based on maps of COV+ location data Additionally it is possible that some communi-ties might get higher rates of false positives because they are located in densely populated areas thus increasing the burden of self-isolation If any of these inequities are identified steps must be taken to mitigate them
Finally numerous actors should engage in the monitoring and eval-uation of DCTT systems Technology developers and public health re-searchers have a clear role in this process Technology developers should work with public health researchers to monitor accuracy precision func-
Ethics of Designing and Using DCTT 47
tionality confidence of estimates sources of error and the like Research-ers may also be able to contribute innovative methods to systematically and rapidly evaluate candidate technologies such as by deploying cluster randomized stepped wedge (Hemming et al 2015) or adaptive trial de-signs and techniques (eg response-adjusted randomization) (Pallmann et al 2018) These approaches were also proposed for use in research to assign candidate experimental treatments and vaccines during the 2014-15 Ebola outbreak (Berry et al 2016) When formal research activities are pursued ethics principles and legal requirements for the conduct of research should apply (eg The Belmont Report)
Furthermore any workplace or institution that incentivizes or man-dates use of DCTT has a responsibility to provide evidence that the in-tervention at minimum is not likely to cause harm and to monitor for unanticipated burdens In all cases it is vital that a trusted intermediary be involved in the evaluation of DCTT programs to limit perceptions of bias and ensure a legitimate basis for decision-making Nonsensitive aggregate DCTT analyses should be made available to the public so as to permit verification and inform continuing public debates about its useful-ness and necessity At an individual level data should be available to users that would permit them to further investigate their personal risk with public health officials or other health workers This is important not only to ensure their health and well-being but also to add a layer of protection against unnecessary quarantine
Recommendations
bull Reviews of DCTT systems must be conducted in part by an inde-pendent intermediary that has established the publicrsquos trust
bull Those who authorize use of DCTT within a particular jurisdic-tion or institution should continuously and systematically monitor the technologyrsquos performance in that context This should include monitoring for effectiveness and benefit monitoring for harms and monitoring for the fair distribution of both benefits and harms They should also monitor evidence that is being generated in other contexts about their selected technological solution and about other competing technologies
48 Digital Contact Tracing for Pandemic Response
bull Data should be available to users that would permit them to further investigate their personal risk with public health officials or other health workers to add a layer of protection against unnecessary quarantine
Public Trust and Public Attitudes
Researchers have estimated perhaps conservatively that DCTT use by 80 of smartphone ownersmdash56 of the populationmdashwill be needed to suppress the epidemic (Hinch et al 2020) These estimates also highlight that some decrease in transmission would be realizable even with lower rates of technology adoption As such in order to maximize impact it is essential to gain a thorough understanding of public perspectives on DCTT including which features and uses of the technology the pub-lic finds acceptable which kinds of DCTT the public would be most likely to use and which designs and uses of DCTT would maintain or jeopardize public confidence and trust There will be variation in public attitudes within and across societies and over time
With respect to what we currently know about public attitudes and trust in DCTT in the United States polling data suggest some potential support and also some divisions regarding willingness to use the technol-ogy Polls conducted by groups based at the University of Zurich (Hargit-tai et al 2020) and the University of Oxford (Altmann et al 2020) suggest that more than 60 of Americans would be willing to install such an app Both a Washington PostndashUniversity of Maryland poll (2020) and a Kaiser Family Foundation poll (Kirzinger et al 2020) show roughly half of the population would be willing to install the app Over half of the population (59) would be willing to share their COVID-19 positive test result with an app in order to anonymously share that information with their contacts (Washington PostndashUMD 2020) Only 29 of respondents to a March 12ndash27 Oliver Wyman Forum poll (Elliott et al 2020) said that they would be willing to share their location data Additionally Washing-ton PostndashUMD data and Pew data from 2019 suggest that approximately one in six Americans do not have a smartphone and thus cannot use the technology without intervention (Pew Research Center 2020)
People may be more willing however to download an app if it will
Ethics of Designing and Using DCTT 49
ease social distancing policies and allow for more economic and social activity Willingness to install a contact tracing app increased among re-spondents to the Kaiser Family Foundation poll from 50 to 66 when respondents were asked if they would be willing to do so to allow schools and businesses to reopen Additionally who develops or administers the app appears to matter Respondents to the Washington PostndashUMD poll indicated higher levels of trust that their anonymity would be preserved by public health agencies and universities than by tech companies or health insurance companies Further more respondents to the Oliver Wyman Forum poll were willing to share their health information with public health authorities (55) than the local government (35) their employer or school (33) or the federal government (27)
These data suggest that people will be more willing to use a contact tracing app when the potential benefits are clearly identified and valued such as lifting social distancing measures and they will be more willing to do so if the data are going to a public health agency rather than the federal government or a tech company Other factors that seem to be as-sociated with greater willingness to install a contact tracing app include younger age and the app source (Hargittai et al 2020) with a preference for apps distributed by public health agencies over others such as health insurers or public universities (Hargittai and Redmiles 2020) However all of this must be read with caution as public polling may not be repre-sentative of some populations or of widespread public attitudes Further these attitudes may shift over time and may be discordant with behaviors (Barth and de Jong 2017)
Deliberative public engagement efforts would be an appropriate means of filling in gaps in understanding about the acceptability of dif-ferent approaches (Fishkin and Laslett 2003 Cavalier 2011) In addi-tion including the public particularly in the earlier stages of planning a path to sustainable resolution to the pandemic could serve to help disseminate a nuanced understanding of what is at stake including the key challenges and trade-offs Aggregated public polling results are not sufficient as a proxy for careful analyses of the ethical challenges but they do provide a necessary input for these analyses Integrating lessons and outputs from public engagement into guidance and other products requires special attention and should be validated and enhanced through further engagement
50 Digital Contact Tracing for Pandemic Response
Recommendations
bull More research into public attitudes is needed In particular in-depth qualitative research should examine public attitudes about percep-tions of trust in DCTT among different communities which fea-tures of DCTT influence trust and the extent to which people are willing to provide different types of data through DCTT to help their community
bull States and localities that are considering adopting DCTT should engage with the public to increase their understanding of the ac-ceptability of DCTT design features and uses among diverse communities
Designing Flexible Technology to Maximize Public Health Utility While Respecting Other Values
Values in Design
Efforts to advance DCTT in the United States and elsewhere have empha-sized the importance of ldquoprivacy by designrdquo that is building privacy and security protections into the design of technology rather than counting on responsible use alone (Cavoukian 2010) As noted above some major technology companies have signaled this position through development of decentralized privacy-preserving proximity tracking (PPPT) systems These systems embed features such as decentralization anonymity of us-ers bans on collection of location data and minimal reliance on or inte-gration of public health authorities or other government actors Many of these features have also been embraced early by advocacy organizations (Crocker Opsahl and Cyphers 2020 Electronic Privacy Information Center 2020 Kahn Gilmor 2020) and in an open letter from nearly 300 researchers (ldquoJoint Statement on Contact Tracingrdquo 2020)
Privacy by design provides principles that incorporate one set of val-ues (privacy) into the design of DCTT Importantly the principles ac-knowledge the need to design privacy defaults into systems while main-taining the capacity of those systems to achieve their otherwise justifiable ends Put another way privacy by design ldquoembraces legitimate non-pri-vacy objectives and accommodates them in an innovative positive-sum mannerrdquo (Cavoukian 2010 p 4)
Ethics of Designing and Using DCTT 51
This stance simple in its statement is not easy to satisfy Given that ldquoobjectivesrdquo are themselves driven by values it begs for an articulation of additional values (aside from privacy) that individuals and groups within societymdashincluding many privacy advocatesmdashmay believe to be important For example at any moment in addition to valuing their own privacy individuals may value efficiency equity autonomy economic well-being companionship patriotism or solidarity Moreover the above stance necessitates an acknowledgment that peoplesrsquo value priorities often change when circumstances change not least of which during a pandemic when mass physical distancing has made it difficult to fully realize many important values (aside from physical privacy) A different orientation is needed at this moment As Flanagan Howe and Nissenbaum (2008) conceptualized in 2008 we should take a ldquovalues in designrdquo approach to DCTTmdashan approach that designs a broader range of values such as those enumerated above into technology
This approach requires a wider ethical lens through which to ex-amine DCTT and requires hard but important work to appropriately balance competing interests within technology architecture For example there is value in technology providing users the option to collect their location history and share it with public health professionals in order to advance the public health response increase system efficiencies (eg by contributing information that can lead to better data processing) and reduce the burden on essential workers For some this might be an ex-pression of autonomy solidarity or patriotism At the same time there is value in further advancing autonomy by designing technology to allow individuals to control what data about them are collected and shared
Justifying a Middle-Ground Approach to DCTT
We ought to embrace a DCTT that has a default of interoperability and privacy protection but that does not stop there Triggering events such as entry of a positive test result or receipt of a notification that one was proximate to someone who tested positive could for example generate a push notification that users can acknowledge in order to permit transmis-sion of potentially useful location data to public health authorities This could be accompanied by an explanation of the value of the information and relevant restrictions on its use
At this point it is worth reiterating that manual contact tracingmdash
52 Digital Contact Tracing for Pandemic Response
which involves collecting information from people whorsquove tested posi-tive and their contactsmdashincludes collection of personal information and potentially embarrassing or sensitive data about the places theyrsquove been and the people theyrsquove had contact with Manual contact tracing efforts use these data to uncover ongoing transmission provide useful informa-tion tailored to the individual and enable isolation and quarantine as necessary
It stands to reason that if these forms of data can be collected by a DCTT and provided to public health authorities in a maximally secure and voluntary way (with clear rules regarding authorized uses) this may amplify public health authoritiesrsquo manual contact tracing efforts For ex-ample location data from DCTT could help jog peoplersquos memories about where theyrsquove been and fill in memory gaps This is especially relevant given the long period of infectivity of SARS-CoV-2 which begins before people are symptomatic and therefore before they are aware they are infected (Ferretti et al 2020) Location data might reveal that a COV+ person was at a restaurant at an exact time and date which could be followed up by contact tracers who could alert the public or use other measures to reach those who were also present in the restaurant at the same time In other disease contexts (see Furlanello et al 2002 Dredze et al 2013 Eckhoff and Tatem 2015 Fraccaro et al 2019) geolocation data have demonstrated some potential to support epidemiology and dis-ease surveillance with technical cautions regarding accuracy and the like (Beukenhorst et al 2017)
These benefits are currently speculative for DCTT At present pro-viding public health authorities with large amounts of data on cases and potential case contacts will be useful only if there is sufficient public health capacity to follow up on these data In addition there is a risk of low-quality data from DCTT flooding the system Investigating poten-tial case contacts identified by a DCTT may distract them from other important efforts and at some point overwhelm public health capacity altogether Whether and to what extent data from DCTT will benefit contact tracing efforts is unknown pointing again to the importance of continuously collecting high-quality evidence about DCTT
Nevertheless what would enable the most flexible and potentially robust public health response is to design DCTT so that restricted data sharing is possible From an ethics perspective the collection and use
Ethics of Designing and Using DCTT 53
of sensitive data in manual contact tracing efforts (described above) is typically seen as ethically justifiable so long as there is sufficient public health benefit and need Thus wouldnrsquot it seem appropriate from both a public health and ethics perspective to design DCTT systems to enable similar data to be shared with public health authorities when and if there is ethical justification for sharing them
Why instead do so many advocate that DCTT should be designed as a ldquominimalrdquo system when this arguably ties the hands of public health and individual users and precludes the collection of data that public health authorities (and indeed many other apps on our phones) typically collect We here consider and appraise some of the reasons that may motivate individuals and groups to argue for minimalistic positions
1 Proponents of minimal systems may believe that such systems will be most
widely adopted Some groups have maintained that only these systems will earn and maintain public trust and be widely adopted (Simpson and Conner 2020) For example the previously referenced open letter (ldquoJoint Statement on Contact Tracingrdquo 2020) asserts ldquoSome of the Bluetooth-based proposals respect the individualrsquos right to privacy whilst others would enable (via mission creep) a form of government or private sector surveillance that would catastrophically hamper trust in and acceptance of such an application by society at large It is crucial that citizens trust the applications in order to produce sufficient uptake to make a difference in tackling the crisis It is vital that in coming out of the current crisis we do not create a tool that enables large scale data collection on the population either now or at a later time Thus solutions which allow reconstructing invasive information about the population should be rejected without further discussionrdquo
Response While it is true that public trust in and acceptance of DCTT is essential for its success there is insufficient evidence that public trust would be threatened by a DCTT system that has the capacity to collect location data and enable voluntary sharing of those data with public health authorities A contrasting perspec-tive is that maintaining public trust requires maintaining public confidence that the DCTT system is providing useful information is benefiting and not harming individuals and is advancing the
54 Digital Contact Tracing for Pandemic Response
public health response (Leprince-Ringuet 2020) From this per-spective a system that is less well integrated into the broader public health response or that generates a higher rate of false positives (as some suggest decentralized approaches might (Fraser et al 2020)) may fare worse when it comes to maintaining public confidence and trust
2 Proponents may hold the view that minimal systems are harmless (or nearly
harmless) to individuals This is because individuals are anonymous none of their location data are gathered and none of their identifiable data are shared with anyone In contrast DCTT systems that collect and share identifiable data including location data may be seen as posing risks of harm to individuals
Response While minimal systems may be harmless (or nearly harm-less) from the perspective of protecting privacy they may not be harmless from the perspective of public health if they generate system inefficiencies through producing too many false positive or false negative contacts Aside from presenting a challenge for public health professionals false positives could also harm individ-uals If users receive a large volume of automated messages alert-ing them to proximity events will this cause distress Will a large volume of alerts cause users to become disengaged and stop using the DCTT or lose confidence in contact tracing more generally as a legitimate method of disease control Admittedly these are just potential harms and risks it is unknown the degree to which they will materialize The point is that privacy-related harms are not the only relevant harms to individuals that we should consider when assessing DCTT
We acknowledge the risk under a middle-ground DCTT of data being used in ethically unjustifiable and harmful ways For exam-ple it would be against the principles and recommendations artic-ulated in this report for data to be sold or monetized by technology companies or others for corporate gain and this misuse of data would be more intrusive if the data were potentially identifiable What makes it ethically justifiable to take this risk is the compen-sating benefit of allowing the most flexible and robust public health
Ethics of Designing and Using DCTT 55
response during the pandemic but this alone is not sufficient The risk of inappropriate uses must be reduced by ensuring stringent requirements for data security and access as well as clear legal protections and recourse for any violations (as discussed further below)
3 Proponents may believe that DCTT systems should not collect location data
as this would be too intrusive and of insufficient value Some proponents of PPPT systems maintain that recording proximity events is sufficient and data relating to usersrsquo movement and location should not be collected (Ingram 2020) The thought may be all we need to know is whether two individuals came into close enough contact for viral transmission to have occurred we donrsquot need to know where or when this contact occurred and there is no need to collect and store usersrsquo location data
Response This conclusion might be too hasty As discussed above there is potential (though unproven) benefit to providing public health authorities with location data Location data could help jog peoplersquos memories about where theyrsquove been provide more context for understanding the nature of ldquoproximity eventsrdquo captured by the DCTT and allow public health authorities to quickly define a category of individuals who may be at risk Collecting location data from cases is what public health authorities do on a regular basis following best practices for manual contact tracing
In addition many peoplersquos location data are currently gathered by apps on their phones and used for various purposes such as to provide more accurate navigation to offer entertainment or to improve services Many are willing to accept these capabilities because they provide some value in return Why not allow DCTT to also collect these data so that the data are available for users to share with public health officials who can then do their work more effectively and refine their understanding of how the disease transmits If many are willing to have these data used to find a bet-ter route home why not let individuals share these data to support the effort to save lives
56 Digital Contact Tracing for Pandemic Response
4 Proponents may hold the view that minimal systems pose little or no threat
to individual autonomy whereas systems that collect identifiable data and
integrate public health do pose a threat to individual autonomy For exam-ple they may worry that use of DCTT could be mandated and not a voluntary choice and in this circumstance mandatory use of minimal DCTT would be less intrusive risky and privacy violating Another worry might be that itrsquos theoretically possible that DCTT could share individualsrsquo data with public health authorities without usersrsquo full understanding if the technology does not even gather identifiable data then itrsquos not possible for these data to be shared without the individualrsquos consent
Response We discuss the importance of appropriately designed disclosures and consent below as well as the high bar that would need to be met to ethically justify mandatory use At this time mandated use of DCTT by states or institutions is not justifiable given uncertainty about potential harms and benefits Users should have a meaningful opportunity to review and understand infor-mation about the specific technology and its uses and to consent Assuming that individuals are not required to use DCTT and that they provide consent to using it designing DCTT to make data collection and sharing possible is the design choice that maximizes individual autonomy because it provides individuals with options they may value
Individuals may wish to share their data with public health au-thorities for both self-interested and altruistic reasons For exam-ple someone who has tested positive for SARS-CoV-2 and enters this test result into an app may wish to be connected to public health authorities in order to be provided with needed information resources and support She may wish for public health authorities to be provided with her phone number in case they need to reach her to provide additional information Further someone who has been alerted by an app that he had a ldquoproximity eventrdquo with a person who has tested positive for SARS-CoV-2 may wish he had location data to share with public health authorities in order to help ascertain whether this event is a cause for concern or whether it is likely a false positive (eg he and the COV+ person were sepa-
Ethics of Designing and Using DCTT 57
rated by a wall) Someone who tests positive for the virus may also wish to share their location history with public health authorities in order to be as helpful as possible to the overall public health re-sponse by facilitating de-identified aggregate analyses that identify locations of higher transmission or contribute to refining overall understanding of the disease and pandemic
5 Concerns about ldquosurveillance creeprdquo and the long-term downstream effects of
digital contact tracing system may also motivate embrace of minimal DCTT Digital contact tracing technology that collects identifiers and loca-tion data and has the capacity to share them with public health au-thorities may represent a massive and concerning increase in govern-ment surveillance of the public It might be feared that the use of this surveillance capacity in the COVID-19 response sets an unwelcome precedent for future use in other contexts Designing DCTT as min-imal systems may be a way to minimize the risk of surveillance creep and to minimize the harms associated with potential future uses of the technology
Response Surveillance creep is a serious concern To guard against surveillance creep protections should be put in place to ensure that only those data that are necessary and relevant for the public health purposes at hand are collected and used and data should be kept only for the period of time needed for those public health purposes In the face of these concerns it is important to emphasize that widespread use of DCTT in the COVID-19 response is justi-fied by the exceptional circumstances of the current pandemic and their use in this context does not imply that future public health use is ethically appropriate without significant public debate (eg use in seasonal flu surveillance efforts) Future use will require in-dependent justification Use of DCTT in other contexts (eg law enforcement or immigration enforcement) is also presumptively unethical
All in all the arguments that DCTT should be designed as a minimal system are not convincing Rather DCTT should be developed through a ldquovalues in designrdquo approach with a core set of features that protect pri-
58 Digital Contact Tracing for Pandemic Response
vacy with enough flexibility to be used differently depending upon local conditions evolving evidence and individual preferences What kind of digital contact tracing system will strike the right balance between public health goals and other considerations will depend upon circumstances For example whether it is even beneficial to provide public health au-thorities with volumes of data about potential contacts of COV+ people will depend in part upon whether they have the capacity to make good use of those data This will vary from location to location and will change over time
Recommendations
bull Technology companies should not alone control the terms condi-tions or capabilities of DCTT nor should they presume to know what may be acceptable to members of the public
bull A ldquovalues in designrdquo approach to development of DCTT should be adopted (Flanagan Howe and Nissenbaum 2008 Knobel and Bowker 2011) Robust public- and user-engagement activities should be pursued to identify and incorporate to the extent pos-sible a range of values into the design of the technology These values may include privacy but also autonomy efficiency equity or others Technology design should reflect an appropriate balance and prioritization of identified values
bull Technology design should not be static but rather it should be ca-pable of evolving depending upon local conditions new evidence and changing preferences and priorities
bull DCTT should be designed to have a base set of features that protect privacy with layers of additional capabilities that users may choose to activate An initial default should be that user location data are not shared but users should be provided with easy mechanisms and prompts to allow for opting-in to this capability with encourage-ment to the public if and as it is shown to be critical to achieving public health goals
Ethics of Designing and Using DCTT 59
Policy Positions to Advance Widespread Use of Digital Contact Tracing Technologies
The public health value of a DCTT depends in part on the number of people who use it This section concerns broad public policy positions that relate to the widespread adoption of DCTT What are ethical means of encouraging or securing widespread adoption of DCTT systems Un-der what circumstances would it be ethical to mandate their use or incen-tivize their use What enforcement challenges exist
Mandating Use
Digital contact tracing has occurred without the publicrsquos explicit volun-tary agreement in some countries such as China and Israel In others use has been voluntary (Valentino-DeVries Singer and Krolik 2020) For example Singapore adopted an app that the public could use on a vol-untary basis and approximately 20 of the population has downloaded and used it Norway has recently launched a contact tracing app that was downloaded by roughly 30 of the population in the first week that it was made available In the United States many advocates and researchers have argued that use of digital contact tracing tools must be fully volun-tary this is the dominant perspective
There are numerous ways that DCTT could be put into use without user choice For example as has been done in Israel location data from mobile phones could be collected and used by the government without usersrsquo consent Use of an app could be formally mandated as a precon-dition for returning to work or school or even further to control entry into a facility or onto transportation such as airplanes through scanning of a QR code to demonstrate personal exposure levels (Gan and Culver 2020)
While these approaches are hard to imagine in the United States some contend that mandatory use of digital contact tracing tools could be ethical and may even be ethically required Mandating use of digital contact tracing tools could in theory vastly increase the effectiveness of digital contact tracing systems and thus may save more lives and allow states to lift lockdowns sooner or avoid reimposing lockdowns in the future Canca (2020) argues that use of privacy-by-design digital contact
60 Digital Contact Tracing for Pandemic Response
tracing tools should be mandatory because the use of these tools will be nearly harmless if there are sufficient privacy protections In addition mandatory use of DCTT that embraces these principles is significantly less intrusive at the individual level than manual contact tracing which involves the collection of personally identifying and potentially sensitive data In this light it could be argued that such mandates are actually pref-erable from the perspective of both public health and individual liberty insofar as they reduce the likelihood of ldquostay at homerdquo orders which are a severe limitation of individual liberty
Nevertheless mandated use of DCTT systems faces considerable ob-stacles For example people may not adhere to the mandate by simply leaving their phone at home thus preventing their activities from being tracked Even more harmful would be if people react to a mandate and a perceived violation of liberty and privacy by employing location and Bluetooth spoofing software to shield their real contacts behind a screen of misinformation The introduction of this misinformation into a contact tracing effort might severely undermine its effectiveness The possibility of nonadherence also raises the issue of enforcement would high rates of nonadherence be permitted or would enforcement be attempted (if even possible) Perhaps more important should the technology not deliver the hoped-for benefits having mandated the use of an unproven technology could result in a loss of public trust in the technology the entity instituting the mandate and potentially the larger public health response (Bernstein et al 2019)
Mandatory DCTT could also be used to enforce quarantine restric-tions and stay-at-home orders for those who are COV+ or are determined to be at heightened risk The use of DCTT in enforcement activities raises a number of ethical (and legal) issues that are beyond the scope of the present analysis In particular individuals have a heightened interest in personal privacy if their data can be used to restrict their freedom of movement and other civil liberties At a minimum stringent procedural protections would be required to ensure that the data collection is fair and unbiased and that DCTT users are provided with adequate informa-tion in advance about how their data may be used
Mandatory use policies for DCTT must therefore convincingly ad-dress a number of questions including
Ethics of Designing and Using DCTT 61
bull Is the technology designed to meet an important and unmet public health need
bull Is there sufficient evidence to suggest that the technology will be effective at serving its purpose
bull Is the outbreak characterized by sufficiently severe morbidity and mortality and a high rate of disease transmission
bull Are there other less autonomy-restricting or less risky alternatives to widespread mandatory use of DCTT
bull Is it possible and likely that a sufficient number of individuals will comply with a mandate
bull Can inequities in the burdens and benefits of the mandate be suffi-ciently addressed through social protections and countermeasures
bull Can enforcement and enforcement discretion be implemented in a manner that is consonant with fundamental rights
bull Will those subject to the mandate interact closely with a population that is at high risk of morbidity or mortality if they contract the virus
bull Is it possible to mandate use and remain consistent with important ethical and legal principles
These questions would need to be satisfactorily addressed and explicitly documented by any decision maker considering mandatory use includ-ing government officials institutional leaders and employers Particularly important is the need to identify reliable evidence that the DCTT would be effective and to ensure that the burdens and benefits of use are equita-ble and justifiable If use of a DCTT is a condition for returning to work or school in person those who refuse or are unable to use DCTT should not lose their jobs or positions as a result and adequate support should be in place for people who are asked to self-quarantine
Finally it is important to distinguish a mandate from a ldquopushedrdquo program installation or a default setting in an application which can be modified by users A mandate relates to a policy of required use whereas the pushed programs or default settings relate to the chosen architecture for download and operation of the application
62 Digital Contact Tracing for Pandemic Response
Incentivizing Use
Perhaps the most effective way to generate widespread adoption of DCTT in the United States is to offer incentives to individuals who choose to adopt and who properly utilize the preferred DCTT approach in a voluntary system External incentives may help ldquonudgerdquo populations toward desired adoption targets Given the importance of widespread use of DCTT modest incentives ought to be considered for DCTT in the US if and when there is sufficient evidence of the technologyrsquos utility Note that in other contexts studies have shown that the provision of some incentive leads to an increase in adoption or utilization of public health programs (Singer and Ye 2013 Lee et al 2014) Moreover even a relatively small incentive can achieve much greater rates of adoption with some studies demonstrating that the incremental adoption gain de-creases as the incentive gets larger (Thornton 2008 Gibson et al 2019) In the context of COVID-19 incentives that might be both effective and ethically acceptable could include a relatively small monetary token free or discounted mobile phone service for a period of time or credit to be used by means of a mobile phone
Not all incentives are ethically appropriate For example making access to lifesaving health care contingent on using a DCTT or making valuable disease information available only to DCTT users would not be ethically appropriate In addition incentives cannot be used to over-come otherwise ethically unjustifiable technology design for example they should not be used as an offset for providing personally identifiable health information to other users
Importantly incentivization schemes must be kept distinct from man-dates as the latter require greater ethical justification To offer an incen-tive is to offer something of actual value to individual participants over and above what they are reasonably entitled to at baseline For example making a return to work contingent on using DCTT is not offering an incentive but instead imposing a mandate and it would have to be justi-fied as a mandate
In the context of COVID-19 it is also necessary to recognize that there is an inherent ldquoincentiverdquo behind the technologymdashthat is the prom-ise of more lives saved faster pandemic recovery and the reduction or elimination of blanket physical distancing Effective public communica-
Ethics of Designing and Using DCTT 63
tion of these goals if and when there is sufficient confidence in the tech-nology is important
Encouraging Use
Another important approach to increasing use of DCTT in the United States is for trusted leaders to encourage their use Community leaders public figures health care professionals and other respected individuals who have the publicrsquos trust and goodwill could be enlisted to commu-nicate with the public about DCTT and encourage its use drawing on notions such as communal responsibility solidarity and so on These en-couragements could be combined with other approaches (eg small in-centives) to optimize reach while continuing to respect individual choice
Recommendations
bull DCTT use should not be mandated at this time given uncertainty about potential harms and benefits Additional technology user and real-world testing is needed
bull Incentives can be a useful complement to encouragements how-ever any incentives for users to install and use DCTT must be equitable should not be coercive and should align with effective use of the technology (ie they should not incentivize downloading an app but then leaving onersquos phone at home)
bull Trusted leaders should be enlisted to communicate effectively with the public about DCTT and encourage its use should the technol-ogy demonstrate some potential The limits of knowledge regard-ing effectiveness should also be explained along with what will be done to improve technological capabilities as understanding evolves
Disclosure and AuthorizationConsent
In deciding whether to use DCTT voluntarily individuals must be suf-ficiently informed both through broad coordinated public engagement campaigns and individual-level disclosures and there must be a meaning-ful mechanism for users to consent It is important to recognize that while
64 Digital Contact Tracing for Pandemic Response
informed consentmdashwhich is characterized by detailed consent forms and requires a witnessed signaturemdashis the standard for most research and clinical care encounters (Faden and Beauchamp 1986) it is not typically the standard for public health disease surveillance In the public health context other relevant protections (such ethics training for public health professionals and strict data handling and confidentiality requirements) are in place and there is a strong public health interest in collecting the relevant data A more limited role for consent has been recommended for public health surveillance based on a reciprocal obligation of members of society to contribute to a ldquocommon goodrdquo and particularly in the con-text of a pandemic practical considerations such as time constraints and exigencies such as increasing morbidity and mortality (WHO 2017)
Under current circumstances given that (1) many individuals have time and capacity to consent (2) DCTT is being considered as part of plans for longer-term restabilization (3) DCTT is not a familiar part of our public lexicon (4) remote consent disclosure and authorization can be easily embedded in DCTT systems (Moore et al 2017) and (5) there are justifiable public deficits in trust with respect to various government and corporate actors handling potentially personal digital information a strong ethical case can be made for requiring a carefully crafted version of what is sometimes referred to as simple consent Simple consent consists of basic disclosure and voluntary agreement or authorization (Ali et al 2017) Three questions then arise
1 What information should be disclosed to potential users of DCTT
bull Information disclosed might include
deg Entity responsible for the technology
deg Its purpose
deg How it works (in lay terms)
Some participatory disease surveillance systems (eg Flu Near You) have received for-
mal ldquowaiversrdquo of consent requirements from institutional review boards (IRBs) in the US
As they undergo development these digital surveillance systems often straddle a line be-
tween public health surveillance and research hence the frequent need or desire to obtain
ethical review by an IRB (Ali et al 2019)
Ethics of Designing and Using DCTT 65
deg What users need to do
deg Any user options eg
Sharing geolocation data with public health authorities when that would facilitate a defined public health goal
Sharing de-identified metadata with technology develop-ers (for system enhancement)
deg User rights
deg How data will be handled
What data are collected
What data are shared (and how and with whom)
Purposes for which data can be used and not used
How data are secured and protected
Whether and what data will be retained (or will be deletable)
deg Potential benefits and any known risks
deg How to obtain answers to questions about the technology and public health response
2 How should this information be presented
Information should be presented leveraging eConsent models that are more accessible than long ldquoclickwraprdquo disclosures typical of mobile apps (Iwaya et al 2019) For example a simple open-source smartphone con-sent module that has been developed by Sage Bionetworks for research uses could be adapted to the public health surveillance context and to DCTT (Doerr Suver and Wilbanks 2016)
bull Formatting recommendations include (cf Doerr et al 2016)
deg simple and straightforward information
deg deliberately organized content
deg multimodal learning (eg visual audio written)
deg accessibility for disabled users
66 Digital Contact Tracing for Pandemic Response
deg multilingual text
deg engagement through interaction (eg swiping to navigate forward and backward)
bull The same simple information should be made publicly available via multiple other platforms (eg on websites in newspapers over social media)
bull More detailed disclosures should be made readily accessible to those who wish to learn more with no hidden surprises
3 How should users signal that they agree to the details specified in disclosures
Opt-in Models
Opt-in models are those that through an affirmative act such as clicking a button users would indicate their intention to use a DCTT This ap-proach is consistent with other app downloads where app details and privacy policies are made available through a download page and users are required to affirmatively click a button to install an app Once in-stalled some apps further alert users to particular ways in which phone capabilities or data will be used with some permitting selective toggling (opting-in or opting-out) of certain features With DCTT apps in addi-tion to disclosures provided on a download page the user could be guided through a simple interactive module embedded in the app (such as is described above) in order to increase the chance of meaningful exposure to important information about the technology and how data will be handled At that point any user options such as those itemized above could be described and choices made
Opt-out Models
There are at least two different ways in which the term ldquoopt-outrdquo has been used in this context The conventional use of the term ldquoopt-outrdquo is characterized by an act which signals an individualrsquos intention to decline something that would have otherwise occurred without intervention A few others have used the term to refer to ldquorevocation of consentrdquo for example the United States COVID-19 Consumer Data Protection Act of 2020 Senate bill (S3663) would establish a default opt-in positionmdash
Ethics of Designing and Using DCTT 67
requiring ldquoaffirmative express consentrdquo for collection and use of prox-imity and other related datamdashand refers to individuals having a right to later revoke their consent through an ldquoopt-outrdquo The latter use of the term is not our focus here
Given this a DCTT app that is voluntarily downloaded through an affirmative act would be difficult to characterize as an opt-out approach This leaves more passive surveillance systems that rely on automatic in-stallation of self-activating technology onto users phones There are a range of views among the authors of this report about the value of an opt-out approach for DCTT with some arguing for an opt-out approach on grounds that it might increase coverage and would be ethically acceptable if accompanied by similar disclosures as above to ensure users are aware of the technology and data uses (Mello and Wang 2020) This approach would present users with a mechanism to opt-out if they wish which should be reasonably easy to effectuate Under these circumstances as noted above an ldquoopt-outrdquo would not be synonymous with mandating use of the technology
Others among the authors argue that there is reason to believe that opt-in approaches may be able to sufficiently achieve desirable levels of utilization relative to opt-out approaches Unfortunately data related to opt-in versus opt-out models of DCTT are very limited One recent sur-vey (Altmann et al 2020) found that across five countries (UK Germany France Italy US) slightly more people reportedly would download an app under an opt-in system (748) than would keep an app on their phone under an opt-out system (677) Moreover when US respondents were directly asked which approach they would prefer 60 indicated a preference for opt-in This remained true across various demographic variablesmdashgender region political affiliation lockdown status and other characteristics Whether actual behaviors would align with anticipated behaviors in the context of DCTT remains an unanswered question that should be carefully studied under real-world conditions There are a range of important empirical questions regarding how much and what kind of impact (positive or negative) various types of defaults might generate for public health and for different mobile phone user groups including vul-nerable and marginalized users
Opt-out models for app authorization may encounter greater legal
68 Digital Contact Tracing for Pandemic Response
and political challenges especially if the COVID-19 Consumer Data Pro-tection Act of 2020 (S3663) the competing Public Health Emergency Pri-vacy Act (S3749) or another similar bill is enacted in the United States Both of these standing bills require affirmative opt-in consent Opt-out approaches also risk negative reactions from some mobile phone users a small number of whom may go so far as to intentionally interfere with data because of the perceived intrusiveness of an automatically installed tracking platform (Dixit 2020)
Given these considerations and the apparent willingness of a large portion of the population to opt-in to use DCTT an opt-in approach to authorization should be instituted to accompany initial DCTT rollout The feasibility and value of opt-out approaches should continue to be evaluated informed by what is technologically possible what local as-sessments of benefits and harms of the technology reveal over time and our evolving understanding of the degree to which an opt-out approach is likely to increase or decrease utilization Opt-out approaches should not be precluded
Recommendations
bull A clear and concise module consisting of basic disclosure and vol-untary authorization should be developed to accompany DCTT This module should not take the form of ldquoclickwraprdquo terms of service or end-user agreements but rather provide only essential information necessary for an individual to make a decision More detailed disclosures (such as FAQs in plain language) should be made easily accessible to those who wish to learn more with no hidden surprises
bull An opt-in approach to authorization should be instituted to ac-company initial DCTT rollout The feasibility and value of opt-out approaches should continue to be evaluated informed by what is technologically possible what local assessments of benefits and harms of the technology reveal over time and our evolving under-standing of the degree to which an opt-out approach is likely to increase or decrease utilization among different populations Opt-out approaches should not be precluded
Ethics of Designing and Using DCTT 69
Promoting Equity and Fairness
Digital contact tracing technologies should be designed and used in ways that as far as possible promote an equitable distribution of benefits and burdens DCTT should be deployed in a manner that does not propa-gate preexisting patterns of unfair disadvantage or distribute harms and risks unfairly throughout the population For example communities with lower rates of technology and data access may benefit less from DCTT Special attention must be paid to communities that experience preexisting health disparities and to those that are being hardest hit by the pandemic
Digital Disparities
In the United States February 2019 data indicate that approximately 80 of the population are smartphone users (Pew Research Center 2020) though rates of mobile phone use are significantly lower among people over age 65 (53) people with any disability (58 2016 data) (Anderson and Perrin 2017) people with less than a high school edu-cation (66) people who earn less than $30000 per year (71) and people who live in rural areas (71) As a result these populations and communities may use DCTT in lower numbers thereby lessening the effectiveness of DCTT and the likelihood of benefit for these populations from such systems Moreover it has been reported that many older and less costly smartphones (roughly estimated at 10ndash20 of smartphones in the US) lack important capabilities required for the leading AppleGoogle platform to work (Bradshaw 2020) This is of special concern because some of the above groups that are less likely to own smartphones in general are also less likely to own newer smartphones with the needed capabilities Some within the above groups (eg people who are older and people identified as Hispanic African American or American Indian) are also disproportionately experiencing morbidity and mortality from COVID-19 (CDC 2020h)
One may argue that by using DCTT human and financial resources that would otherwise be spent on manual contact tracing will be pre-served and these resources can then be redirected to better meet the needs of those who are not otherwise being effectively served by the technology because of disparities or for other reasons This argument has intuitive
70 Digital Contact Tracing for Pandemic Response
appeal and should be taken seriously however it is unsettled whether DCTT will contribute sufficient efficiencies to the overall public health response to make it possible financially and logistically for manual ser-vices to be allocated in greater proportion to those who are unable to benefit from DCTT It is entirely possible that at least in the short-term DCTT may introduce new inefficiencies due to unintended consequences or the need for public health officials to follow up many more contacts One possible mitigation to the challenge of digital disparitymdashthough it does not solve the underlying challenge of ensuring net efficiency across systemsmdashmight be to provide mobile phones or other devices and data packages to those who would otherwise be left out
Disparate Risk of Harm from Surveillance and Data Gathering
Ensuring wide digital coverage does not however resolve other equity concerns It is important to consider that some populations may experi-ence greater harm and fear of harm from having their data collected For example some groups such as African Americans Hispanic Americans Muslim Americans and undocumented immigrants have more reasonable fear of their data being handed over to law or immigration enforcement and some groups have lower levels of trust in public health due to past injustices (CSM 2017 Pew Research Center 2017 Rodrigues et al 2018 Auxier et al 2019) Any data gathered by DCTT should be used solely for public health purposes Efforts should be made to assure members of these and other communities that their data will not be misused or made available to those outside of a public health context In addition if DCTT are used in the current pandemic this should be with the understanding that future use of DCTT in other contexts (eg law enforcement or im-migration enforcement) is presumptively unethical
Some preliminary polling related specifically to DCTT emphasizes the complexity of the challenges faced and the need for deeper public engagement (Anderson and Auxier 2020) The polling results suggest that people who identify as African American or Hispanic are more likely than people who identify as White to consider government tracking of mobile phones as acceptable These findings like many others are difficult to in-terpret given background political polarization on the issue More direct engagement is required to better understand how different communities comprehend and experience DCTT and other forms of surveillance
Ethics of Designing and Using DCTT 71
Discrimination and Stigma
Stigma may result from an individual being identified as COV+ or a neighborhood or establishment becoming identified as a ldquohotspotrdquo as a result of numerous COV+ people living in that area or having visited that establishment In particular certain groups may suffer more as a result of being associated with COVID-19 such as the well-documented blame that has been directed toward Chinese people (and broadly East Asian communities) or the communities that are disproportionately likely to contract the illness (Devakumar et al 2020) When identifiable lo-cation data are made public as has been the case in South Korea per-sonal and private information were revealed Furthermore businesses in South Korea that were identified as having patrons who tested positive for COVID-19 have suffered economic losses and stigma (N Kim 2020)
To avoid the stigma and potential discrimination that can result from being identified as COV+ DCTT must never make data publicly avail-able that could be used to identify persons who have tested positive Safe-guards must be in place to ensure that any identifiable data that may be gathered for public health purposes are protected If DCTT data are used to provide heat maps to the public of locations that COV+ individuals frequently visit so as to provide representations of geographic risk or for other reasons it is essential that care be taken to avoid unfairly distrib-uting further economic burdens or other stigmatizing and discriminatory outcomes
Recommendations
bull A commitment to equity means a commitment to ensuring that the benefits and burdens of DCTT are distributed fairly Public engagement is an important tool for assessing impact and to rectify inequities
bull States localities and institutions that recommend widespread use of DCTT should provide technology (eg mobile phones Blue-tooth devices) and free data packages to those who desire but lack access to these devices
bull If there are lower rates of adoption of DCTT systems in some identifiable communities public health authorities should iden-
72 Digital Contact Tracing for Pandemic Response
tify ways to compensate For example directing more non-DCTT resources and efforts toward those communities to meet specific needs that are elsewhere being supported by technology
bull If maps are generated based on DCTT to provide the public with the locations that COV+ individuals have visited steps must be taken to minimize the stigma and potential financial losses that could result from being identified as a hotspot
Instituting Transparent Governance and Oversight
DCTT must be developed with an eye toward both present and future implications Transparent and publicly trustworthy management gover-nance and oversight of DCTT technology and data is both a near- and long-term necessity We face significant uncertainties DCTT technologies are rapidly developing Their risks capabilities effectiveness and down-stream implications are not yet well understood
Concerns about ldquoSurveillance Creeprdquo
Significant concerns have been expressed by privacy advocates (Guari-glia 2020) and in the popular press (Giglio 2020) about what is known as ldquosurveillance creeprdquo Their worry is that state and corporate actors will use new surveillance technologies capacities and permissions well beyond the purposes for which they were initially justified to the public and beyond the time when they are useful for the COVID-19 pandemic
Surveillance creep should be guarded against Only those data that are necessary and relevant for the public health purposes at hand should be collected and used and data should be kept only for the period of time needed for those public health purposes Data should be used only for public health purposes
Any use of DCTT during the current pandemic would be justified by the circumstances of this pandemic and its use in this context does not set a precedent for future public health use (eg use in seasonal flu surveillance efforts) Future use will require independent justification Use of DCTT in the future in other contexts (eg law enforcement or immi-gration enforcement) is presumptively unethical
Broadly speaking efforts should be made to generate public aware-
Ethics of Designing and Using DCTT 73
ness and consensus that use of DCTT in COVID-19 efforts does not imply that future use is justifiable However generating this public aware-ness may be particularly challenging given the complexity of the informa-tional environment where public debate ranges from legitimate concerns about surveillance creep to conspiracy theories regarding the origins of the COVID-19 pandemic (Muller 2020) This means authorities bear spe-cial obligations to be clear on how they plan to use the technologies what oversight mechanisms will be employed to address potential abuse and how they intend to publicize the conditions under which programs will be terminated making sure they are followed
Oversight and Ethical Review
We are rapidly gaining knowledge about SARS-CoV-2 and COVID-19 but we still have essential gaps in our understanding In the United States public health responses including DCTT will generally be developed and coordinated by individual states regional consortia and associations (Reston Sgueglia and Mossburg 2020) Good governance in this con-text requires transparency and the creation of oversight bodies with the appropriate expertise and representation to allow nimble and effective responses while serving as trusted representatives
To address the range of ethics-related concerns about the design and use of DCTT digital surveillance oversight committees should be estab-lished perhaps at a state level and with a platform for national coordina-tion These committees can provide ethical and regulatory review prior to and concurrent with widespread use of DCTT These committees should be composed of a diverse group of experts capable of evaluating a DCTT system locally including members of communities that experience higher rates of digital disparity
When assessing the design and use of digital contact tracing systems these committees (and the public more widely) should consider not only the risks and benefits accrued during the COVID-19 pandemic but also implications for the future What kind of precedent might use of these technologies during the current pandemic set for future use capabilities in other infectious disease outbreaks or in other social contexts (eg law enforcement) How can we navigate safe use of these technologies in a way that preserves public trust in them and enables the possibility of future beneficial use
74 Digital Contact Tracing for Pandemic Response
As a start it should be emphasized that principles offered in this and other guidance documents do not apply only during the pandemic Future efforts to advance DCTT capabilities during quieter times should make every effort to follow them
Recommendations
bull Digital surveillance oversight committees should be established expeditiously with diverse and qualified membership to provide ethical and regulatory review prior to and concurrent with wide-spread use of a DCTT system
bull Understandable and publicly accessible rules must guide the col-lection access control use storage and combination of data by government authorities public and private institutions and other parties such as public health researchers
bull Only those data that are necessary and relevant for the public health response to COVID-19 should be collected and used
bull Identifiable data should be kept only for the period of time needed for the public health response to COVID-19
bull Identifiable data collected as part of this response should not be shared with anyone other than the relevant public health authorities without additional specific informed consent of individual users
bull Before a government or institution adopts a digital contact tracing program they should state the conditions under which the digital contact tracing program will be terminated
bull Future use of DCTT to advance public health or other efforts (eg use in seasonal flu surveillance) would require independent justification DCTT designed for public health use should not be used by law or immigration enforcement
bull The principles offered in this guidance document apply both during and following the COVID-19 pandemic
75
The implementation of digital contact tracing technology (DCTT) is likely to implicate a number of US laws at both the federal and state levels This section focuses primarily on federal laws as these laws apply nationwide and generally preempt conflicting state laws A comprehen-sive assessment of the legality of any particular DCTT program would require case-specific analysis and attention to relevant state laws includ-ing any that specifically address DCTT which may soon exist in one or more states The analysis here is limited to the United States foreign and international laws will not be addressed
Many of the laws discussed in this section are privacy laws designed to protect individuals from the harms that may result from the unautho-rized or improper use of their personally identifiable information (PII) Under these laws legal concerns will generally be minimized if privacy protections are built directly into the DCTT technology (eg ldquoprivacy by designrdquo) As a general principle DCTT should be designed to collect and store only as much PII as is necessary to achieve the public health purpose Collecting only proximity data for example is likely to raise fewer legal concerns than collecting both proximity data and geolocation data Likewise creating aggregated anonymized or de-identified data will raise fewer legal concerns than using and disclosing PII
As we have argued elsewhere in this guidance document however the public health and societal crisis caused by COVID-19 may justify
Legal Considerations
FOUR
76 Digital Contact Tracing for Pandemic Response
greater encroachments on individual privacy than would otherwise be permissible Regardless of the type of data collected privacy concerns will be reduced if users are afforded the right to choose whether their PII is collected and how it is used and disclosed As such DCTT should gener-ally secure meaningful user consent before collecting PII a process which typically requires both disclosure of relevant information and agreement on the part of the user
Privacy concerns will also be reduced if the use of PII is strictly lim-ited to tracking and limiting the spread of SARS-CoV-2 The use of DCTT data for other purposesmdashsuch as commercial or law enforcement pur-posesmdashwould raise additional legal and ethical concerns In addition DCTT developers may be required to implement governance policies that ensure the secure storage of PII limit data retention periods require transparency about data sharing and maintain records of responses to data requests from government authorities
In short the legality of a DCTT program under current United States law will depend on a number of factors including what type of data is collected how the data are used and who may access them how user consent is obtained whether the entity collecting and using the data is the government or a private corporation the context in which data are collected (eg employment education or commercial) and which states have jurisdiction over the program
Privacy law in the United States unlike in other jurisdictions such as the European Union (EU) and Australia is generally sector-specific and limited in scope The result is a patchwork of protections that differ significantly depending on the entity that collects the data and the type of data collected For example under current law telecommunication carriers are governed by different privacy rules than mobile broadband providers Given the complexity of existing federal privacy law we be-lieve that it would be beneficial for the US Congress to enact new privacy legislation that is specifically tailored to the use of DCTT in response to COVID-19 Congress appears poised at least to debate such legislation a pair of bills recently introduced in the Senate and one in the House of Representatives would significantly restrict the collection of PII by digital devices for COVID-tracing purposes S3663 S3749 HR 6866 116th Cong (2020)
Legal Considerations 77
Data Privacy and Data Security Laws
Telecommunications
A DCTT provider that collects data from a userrsquos mobile phone may be subject to the privacy rules governing telecommunication carriers which are enforced by the Federal Communications Commission (FCC) The data protected under these rules are limited however to certain types of PII termed ldquocustomer proprietary network informationrdquo (CPNI) More-over the rules generally apply only to telecommunications carriers and interconnected VoIP (Voice over Internet Protocol) providers
In particular under section 222 of the Communications Act of 1934 47 USC sect 222 and the implementing regulations of the Federal Com-munications Commission (FCC) telecommunications carriers and VoIP providers must establish and maintain systems designed to ensure that they adequately protect their subscribersrsquo CPNI and they are generally restricted from using or disclosing CPNI without the customerrsquos consent (unless the use of disclosure is needed to provide the services subscribed to by the customer) If customer consent is sought to use or disclose CPNI individual notice must be provided to the customer and such notice must provide sufficient information to enable the customer to make an in-formed decision as to whether to permit the requested use or disclosure
CPNI is individually identifiable information that carriers and pro-viders have collected about their customers including phone numbers called and the frequency duration and timing of such calls Of most relevance to DCTT a recent FCC Notice of Apparent Liability asserted that user geolocation data collected by mobile phone network carriers qualify as CPNI under sect 222 and related rules 35 FCC Rcd 1785 (2) (2020) Pursuant to this notice the FCC fined T-Mobile for selling to third parties location data that were derived from the communication between the mobile phones of T-Mobilersquos customers and nearby network signal towers (The FCC also levied fines against ATampT Verizon and Sprint on the same grounds (Valentino-DeVries 2020)) While the FCC has made its position clear that geolocation data are CPNI courts have yet to weigh in on the matter
Even if geolocation data are CPNI however the FCC can enforce sect 222 of the Communications Act only against telecom carriers and VoIP
78 Digital Contact Tracing for Pandemic Response
providers not against cable broadband and mobile broadband internet providers 47 USC sect 53(44) 47 CFR sect 93 In 2018 the FCC promul-gated a regulation stating that contrary to its prior position its sect 222 authority does not extend to cable broadband and mobile broadband internet providers Restoring Internet Freedom 83 Fed Reg 7852 (Feb 2 2020) (to be codified at 47 CFR pts 1 8 and 20) This regulatory shift was subsequently upheld by the DC Circuit Mozilla Corporation v Federal Communications Commission 940 F3d 1 (2019)
In addition to sect 222 the FCC has authority to regulate ldquocommon carriersrdquomdashincluding both telecommunication carriers and broadband internet providersmdashunder sect 201(b) of the Communications Act In the past the FCC has interpreted sect 201(b) to protect against ldquounjust and unreasonablerdquo privacy and data security practices with respect to custom-ersrsquo personal information beyond CPNI In 2016 the FCC promulgated a regulation asserting its authority under this interpretation However Congress overturned this regulation pursuant to the Congressional Re-view Act in 2017 SJ Res 34 115th Cong (2017) At present the extent of the FCCrsquos authority under sect 201(b) remains unsettled (Mulligan and Linebaugh 2019)
Consumer Protection
The collection storage release and transmission of digital user data in-cluding proximity contacts is more generally governed by the Federal Trade Commission (FTC) The FTC is an independent US law enforce-ment agency tasked with protecting consumers and promoting competi-tion across broad sectors of the economy (FTC 2020) The FTCrsquos primary legal authority with respect to consumer protection comes from Section 5 of the FTC Act which prohibits ldquounfair or deceptive acts or practices in or affecting commercerdquo 15 USC sect 45(a)(1) Note that the FTC and FCC have some overlapping authority to protect consumer privacy in the context of telecommunications (FCC and FTC 2017)
The FTC has interpreted Section 5 to require companies to be trans-parent and accurate about their collection of PII from consumers A com-pany may be found to have engaged in a deceptive practice if it fails to disclose that it is collecting user data or fails to disclose that it is sharing these data with third parties and to provide a general description of these third parties The FTC has used its authority under Section 5 numerous
Legal Considerations 79
times to discipline companies that purport in published privacy policies or other notices to provide protection for the privacy andor security of personal information yet fail to do so in practice For example the FTC may find it both ldquounfairrdquo and ldquodeceptiverdquo for a mobile app privacy policy to state that the app never discloses location information to third parties when in fact the app shares that information with the app developerrsquos service provider which in turn uses it to provide analytical data to the app developer that are used to create targeted advertising
The FTC does not use its Section 5 authority other than to protect consumers and generally does not consider ldquode-identifiedrdquo user data which are data that are not ldquoreasonably linkablerdquo to a consumer to be a subject for consumer protection In general data collected are not ldquorea-sonably linkablerdquo so long as the company collecting it ldquo(1) takes rea-sonable measures to ensure that the data are de-identified (2) publicly commits not to try to reidentify the data and (3) contractually prohibits downstream recipients from trying to reidentify the datardquo (FTC 2012)
Many states have laws that are similar to Section 5 prohibiting un-fair and deceptive acts and practices Both Section 5 and these similar state laws can be violated not only by misrepresentations (affirmative deception) but also by material omissions Thus a failure to inform an app user of the apprsquos collection of tracking data and the planned use and disclosure of those data could constitute a violation of these laws Com-panies providing DCTT apps should make sure that all such information is disclosed in the appsrsquo terms of use to which users must affirmatively agree
Childrenrsquos Online Privacy
Children who use DCTT may be protected by additional privacy protec-tions In particular collection of digital PII from children under the age of 13 is strictly regulated under the Childrenrsquos Online Privacy Protection Act (COPPA) (15 USC sectsect 6501ndash6505) Under COPPA PII includes ldquofirst and last name[] a persistent identifier that can be used to recognize a user over time and across different online services[] and geolocation infor-mation sufficient to identify street name and name of a city or town[]rdquo COPPA prohibits a website or online service from collecting personal information (including location information) from children under age 13 without obtaining verifiable parental consent Note that there may be an
80 Digital Contact Tracing for Pandemic Response
exception to this requirement for an ldquoinvestigation on a matter related to public safetyrdquo 16 CFR sect 3125(c)(6)(iv)
Electronic Surveillance
In addition to misuse of user data by DCTT providers another privacy concern is that a third party may be able to access sensitive PII that is collected and stored by a DCTT system without the userrsquos knowledge and consent There are a number of federal criminal laws however that would likely prohibit such unauthorized access to PII
In particular the Electronic Communications Privacy Act of 1986 (ECPA)mdashwhich includes the Wiretap Act (18 USC sectsect 2510ndash2522) the Stored Communications Act (18 USC sectsect 2701ndash2711) and the Pen Register Act (18 USC sectsect 3121ndash3127)mdashmakes it a crime to access elec-tronic communications without authorization Individuals who violate the ECPA face up to five years in prison and fines up to $250000 Victims are also entitled to bring civil suits and recover actual damages in addi-tion to punitive damages and attorneyrsquos fees for violations
Generally the access restrictions in the ECPA apply unless consent is given or if access is authorized by statute for law enforcement purposes For example an employer is generally forbidden from accessing an em-ployeersquos private emails However if consent is given in the form of an employment contract that explicitly authorizes the employer to access emails it may be lawful under the ECPA for the employer to access such information Along the same lines the ECPA would likely prohibit an employer from accessing contact tracing data on an employeersquos phone without the employeersquos consent However the ECPA would likely not prohibit duly authorized government public health officials from access-ing contact tracing data without consent
As its name suggests the Stored Communications Act (SCA) regu-lates access to communications at rest that is not in transit The SCA makes it unlawful to intentionally access a facility in which electronic communication services are provided and to obtain alter or prevent au-thorized access to a wire or electronic communication while it is in elec-tronic storage in such a system As such the SCA would likely apply only to centralized collection of contact tracing data
The Pen Register Act covers any ldquosignaling informationrdquo exchanged in a communication such as phone numbers The statute does not reach
Legal Considerations 81
the content of such communications however An expansive interpreta-tion of the Pen Register Act would cover Bluetooth ldquohandshakesrdquo as they are merely signaling information between devices which do not carry content See United States v Forrester 512 F3d 500 (9th Cir 2007) (find-ing that email headers and IP addresses are akin to pen registers and have no Fourth Amendment protection) Unlike the SCA there is no statutory exclusionary rule that applies when the government illegally uses a pen register trap and trace device Additionally there is no private cause of action against the government for violations of the Pen Register Act
State Data Privacy Laws
States have a variety of privacy laws and are increasingly seeking to reg-ulate the online collection of personal information and the use and dis-closure of such information To date most of these laws focus more on transparency and protection from unauthorized access than on restricted collection and use (except with respect to biometric information) seek-ing to ensure that individuals who use websites or online services such as mobile applications do so on an informed basis with respect to the privacy provided by those sites and services Two examples of such state laws are the California Online Privacy Protection Act (CalOPPA) and the California Consumer Privacy Act (CCPA) Both laws require notice to in-dividuals who use websites or online services such as mobile applications in order to ensure that users are informed about the privacy of personal information collected by those sites and services (The CCPA also applies to data collection off-line) Both laws treat IP addresses and location data as types of potentially identifiable personal data and so would very likely apply to DCTT apps used by California residents
CalOPPA requires that the operator of any website mobile appli-cation or other online service (ldquoSiterdquo) post a privacy policy on the Site disclosing certain information regarding the Sitersquos collection use and dis-closure of PII CalOPPA applies to any Site that is accessible to California residents The required disclosures are not onerous and would apply only to collection of data that are identifiable to an individual person (but depending on who collects the data location data together with a device identifier are identifiable to the user)
The CCPA requires that any entity qualifying as a ldquobusinessrdquo provide its ldquoconsumersrdquomdashdefined as lawful residents of Californiamdashwith specific
82 Digital Contact Tracing for Pandemic Response
disclosures about the businessrsquos collection use and disclosure of personal information Importantly the CCPA applies only to for-profit businesses that meet certain thresholds of revenue or access to consumer informa-tion A public health agency or a nonprofit organization would not be subject to the CCPA Cal Civ Code sect 1798140(c)
The CCPA defines ldquopersonal informationrdquo as ldquoinformation that iden-tifies relates to describes is reasonably capable of being associated with or could reasonably be linked directly or indirectly with a particular consumer or householdrdquo The statute provides a nonexclusive list of po-tential identifiable personal information including ldquogeolocation datardquo In accordance with the CCPA businesses must provide consumers with a notice ldquoat or before the point of collectionrdquo of personal information which must describe the personal information to be collected and the pur-poses for collecting that information Businesses must additionally allow consumers to request access to and request deletion of personal informa-tion Businesses must allow for consumers to opt-out of the sale of any personal information Developers of COVID-tracing apps would want to build in compliance with these requirements In addition California Civil Code sect 1798815(a)(1) requires companies to ldquomaintain reason-able security procedures and practices appropriate to the nature of the information it processesrdquo
Like privacy laws generally the CCPA does not grant consumers rights regarding the use of de-identified information However the CCPA does require businesses to implement processes that prohibit re-identification of de-identified information as well as technical safeguards to prevent inadvertent release of that information Cal Civ Code sect 1798140(h)
Health Information Privacy
Many DCTT systems will be designed to collect health-related data of users such as symptom tracking SARS-CoV-2 test results and prior ex-posure to a person who is COV+ Individuals may have additional privacy protections with respect to the use and disclosure of this health-related information
The use and disclosure of individually identifiable health information is strictly regulated under the privacy and security rules implementing the
Legal Considerations 83
Health Insurance Portability and Accountability Act (HIPAA) HIPAA is limited in application however to health care providers and health insur-ance plans (ldquocovered entitiesrdquo) and ldquobusiness associatesrdquo of such entities ldquoBusiness associatesrdquo under HIPAA are persons who perform services for covered entities and need access to personal health information to do so
HIPAA-covered entities must have written authorization to use or disclose identifiable health information (ldquoprotected health informationrdquo or PHI) from the individual to whom such information pertains unless the HIPAA regulations promulgated by the US Department of Health and Human Services (HHS) provide an exception to the requirement for such individual authorization
Among the exceptions to the individual authorization requirement is an exception for certain uses and disclosures of PHI for public health purposes 45 CFR sect 164512(b) This exception would permit for exam-ple a HIPAA-covered entity to disclose the PHI of an individual who tests positive for SARS-CoV-2 to a public health authority A ldquopublic health authorityrdquo is an agency or authority of the US government a state ter-ritory a political subdivision of a state or territory or Indian tribe that is responsible for public health matters as part of its official mandate as well as a person or entity acting under a grant of authority from or under a contract with a public health agency such as a contact tracer Id sect 164501
Many DCTT developers are HIPAA business associates and any use and disclosure of PHI collected through DCTT used on behalf of HI-PAA-covered entities is restricted under the HIPAA privacy rules Nota-bly in response to COVID-19 HHS announced that its Office for Civil Rights would exercise its enforcement discretion and would not impose penalties for violations of certain provisions of the HIPAA Privacy Rule against health care providers or their business associates for the good faith uses and disclosures of protected health information for public health and health oversight activities during the nationwide public health emergency 85 FR 19392 (2020)
Many states also have health information privacy laws The HIPAA privacy rule sets a ldquofloorrdquo of privacy protections allowing the states to be more protective of privacy More specifically HIPAA preempts a state law if (but only if) the state law is ldquocontraryrdquo and less protective of privacy than the HIPAA privacy rule However if a state law is determined by the
84 Digital Contact Tracing for Pandemic Response
Secretary of HSS to be necessary to serve a ldquocompelling need related to public health safety or welfarerdquo it may survive preemption even if it is less privacy-protective than HIPAA 45 CFR sect 160203 (a)(1)(iv)
The Public Health Service Act also restricts the use of certain per-sonally identifiable information collected by entities involved with public health activities without the individualrsquos consent 42 USC 242m(d)
Labor and Employment Privacy Rights
Labor and employment lawsmdashthat is laws that govern the relationships between employers and employeesmdashmay prove relevant to DCTT espe-cially if employers mandate the use of DCTT or seek to collect health information regarding their employees using DCTT Depending on the built-in privacy protections of the DCTT system an employer may be able to access important health information from an employeersquos phone As noted above the ECPA would generally prohibit an employer from ac-cessing this information without the employeersquos consent Even with con-sent however there are limits on the collection and use of an employeersquos health information
In particular the use of DCTT may raise special concerns about em-ployment discrimination for example if an employer were to fire an em-ployee who tests positive for SARS-CoV-2 (COV+) or who has a known SARS-CoV-2 exposure The Americans with Disabilities Act (ADA) pro-tects disabled employees from discrimination and restricts the collection of personal health information by employers The Equal Employment Opportunity Commission (EEOC) which is the federal agency tasked with enforcing the ADA in the employment context would likely con-sider COV+ to be a ldquodisabilityrdquo under the ADA and analogous state laws prohibiting discrimination against disabled people COV+ is likely to be a ldquodisabilityrdquo especially where the individual is symptomatic andor experi-ences related health issues or if it is later determined that testing positive for SARS-CoV-2 leads to long-term or chronic health effects ldquoExposure to a COV+ personrdquo could also be covered by those laws because a person exposed to a COV+ individual could well be perceived as being disabled by being considered likely to be infected
Legal Considerations 85
The ADA generally requires that businesses make ldquoreasonable accom-modationsrdquo for persons who are disabled which may include individuals who are COV+ or who have a preexisting disability that places them at higher risk from or may be exacerbated by COVID-19 The EEOC has published guidance on reasonable accommodations under the ADA and related laws in the context of COVID-19 (EEOC 2020) Among other things this guidance clarifies that consistent with the ADA employers may take temperatures or otherwise collect health information about employees during the pandemic crisis so long as they keep that infor-mation confidential As of May 18 2020 the EEOC has not provided guidance that specifically addresses the applicability of the ADA to the use of DCTT by employers
In addition employment laws such as the ADA and the Family and Medical Leave Act (FMLA) and state law equivalents generally limit disclosure of information and require employers to keep confidential any employee personal health information related to a disability or request for medical leave Under the ADA any information regarding the medical condition or history of an employee that an employer obtains as part of an examination or inquiry into a disability could constitute a confidential medical record that can be disclosed only to certain individuals in lim-ited circumstances 42 USC sectsect 12112(d)(3)(B) and 12112(d)(4) The FMLA also prevents the disclosure of records related to medical histories in connection with an employeersquos leave request or eligibility 29 CFR sect 825500(g) The EEOC and some courts have gone further and taken the position that any information concerning an employeersquos medical con-dition is protected under the ADA or FMLA
As discussed elsewhere in this guidance document employers may have a good reason to employ DCTT in order to ensure workplace safety and limit the spread of SARS-CoV-2 in the community Employers may also face legal liability if they fail to protect employees (or customers) from potential exposure or infection In particular employers have an obligation under the Occupational Safety and Health Act to keep the workplace safe for employees In response to COVID-19 the Occupa-tional Safety and Health Administration (OSHA) has developed guid-ance on preparing the workplace (OSHA 2020) The CDC has also pre-pared guidance on healthy business operations and reducing the spread
86 Digital Contact Tracing for Pandemic Response
of SARS-CoV-2 in the workplace (CDC 2020c) Employers must strike an appropriate balance between avoiding employment discrimination and promoting workplace safety
Reflecting the need for such a balance the employee protections un-der the ADA and other employment laws are not absolute and are limited by among other things the need to protect the health and safety of other employees and the public Protection for workplace safety and health generally will justify appropriately tailored measures such as inquiries into an employeersquos personal health status or whether someone has tested positive for SARS-CoV-2 temperature checks and removal of employees from the workplace who are experiencing symptoms or have tested posi-tive and have not been cleared to return to work
Note finally that the use of DCTT by employers should be evaluated in conjunction with the hazard pay sick leave and other benefits that are available to employees Under the Families First Coronavirus Response Act employers with more than 50 employees and fewer than 500 employ-ees are required to provide two weeks of paid sick leave to an employee who stays home because of COVID-19 Pub L No 116-127 134 Stat 178 (2020) This paid leave extends to those who are themselves ill are quarantined or are awaiting a diagnosis as well as those who are caring for sick family members However reporting suggests that more than 75 of US workers will not qualify for benefits under this act (Cochrane Miller and Tankersley 2020)
Constitutional Privacy Rights
A DCTT program involving only private actors operating on the ba-sis of voluntarily provided information would not present constitutional privacy issues But any government-directed use of digital technology to support public health tracking and contact tracing involving mandatory government surveillance may potentially implicate a variety of consti-tutional protections These constitutional protections apply to actions taken by any level of government in the United States While state gov-ernments have broad policing powers in the area of public health (Jacob-son v Massachusetts 197 US 11 (1905)) and are generally allowed to enforce legislation not preempted by federal laws even emergency and
Legal Considerations 87
health-protective laws must be consistent with the US Constitution (HHS 2019 CDC 2020f)
Fourth Amendment Search and Seizure
Many people considering whether to use a DCTT app may be concerned that government enforcement agencies would obtain tracing data and use those data to conduct criminal prosecutions or immigration proceed-ings Constitutional protections notably the Fourth Amendmentrsquos limit on warrantless searches limit the governmentrsquos use of personal data in the criminal context However exceptions exist allowing law enforce-ment to access information even when such access would generally be prohibited How the government accesses personal data stemming from contact tracing needs to be scrutinized and protections will hinge on the manner of access
In general the Fourth Amendment protects ldquo[t]he right of the peo-ple to be secure in their persons houses papers and effects against un-reasonable searches and seizuresrdquo As originally interpreted the Fourth Amendment was considered tied to common-law trespass That is no lon-ger the case US Supreme Court precedent interprets the Fourth Amend-ment to protect ldquopeople not placesrdquo and extends to the protection of certain expectations of privacy such as location information as long as such expectations are reasonable Katz v United States 389 US 347 351 (1967) A warrantless government search is unconstitutional when the information sought is private and such expectation of privacy is ldquoone that society is prepared to recognize as reasonablerdquo Smith v Maryland 442 US 735 743ndash44 (1979)
The constitutionality of a search will revolve around the following analysis whether the digital program either violates an individualrsquos ldquorea-sonable expectation of privacyrdquo (likely triggered by programs collecting large amounts of location andor health data) or involves a government ldquotrespassrdquo (likely triggered by required app downloads) Katz v United States 389 US 347 (1967) United States v Jones 565 US 400 (2012)
Courts will most likely weigh the intrusiveness of the measures taken in implementing a search standard against the severity of the situation governmental and individual interests and accountability measures and safeguards built into the system
Voluntary sharing by individuals of their information with other par-
88 Digital Contact Tracing for Pandemic Response
ties including the government would mean that there was no reasonable expectation of privacy and would not raise the issues elaborated above It is worth noting that consent may not be considered voluntary if coerced or conditioned especially with regard to public employees or students of public institutions
Third-Party Doctrine
Some legal doctrines allow for the governmentrsquos acquisition of otherwise private information consistent with Fourth Amendment privacy protec-tions The third-party doctrine for example provides that individuals have no reasonable expectation of privacy in information voluntarily shared with others even if the information is revealed on the assumption that it will be used only for a limited purpose and the confidence placed in the third party will not be betrayed Smith v Maryland 442 US 735 (1979) United States v Miller 425 US 435 (1976) This applies to in-formation provided by third parties (mobile carriers internet service pro-viders medical tracking device manufacturers etc) to the government under order or request even when the third partyrsquos end-user agreements or privacy policies create an expectation of privacy
The Supreme Court has narrowed the applicability of the third-party doctrine to exclude use and disclosure of ldquohistoricalrdquo cell-site location information (CSLI) data For example in Carpenter v United States 138 S Ct 2206 (2018) the Court reasoned that the third-party doctrine does not justify use and disclosure of historical CSLI because an individual does not provide that information voluntarily Rather that information is pervasively collected by the cell phone company without any affirmative action on the part of the individual The Court did not express a view on ldquoreal-timerdquo CSLImdashlocation information that live-tracks a cell phonersquos locationmdashor on GPS data that may be stored in the phone itself
The Carpenter decision builds on a line of cases related to searches of digitally stored location data In Riley v California 134 S Ct 2473 (2014) the Court held that absent exigent circumstances law enforce-ment must obtain a warrant to search an individualrsquos phone Exigent circumstances are those that require immediate action because there is a probability that evidence may be destroyed The use of a centralized data-base for collection of digital contact tracing data would obviate deletion
Legal Considerations 89
concerns If the data are stored locally in the phone issues may arise as to whether law enforcement may suspect the data may be deleted following an arrest
Similarly in United States v Jones 132 S Ct 945 (2012) Justice So-nia Sotomayor authored a concurring opinion arguing that the use of a GPS to track a defendantrsquos whereabouts has the potential of providing the government with enough data points to create a ldquomosaicrdquo of the personrsquos life Location data obtained through centralized location contact tracing have the potential of providing information on an individualrsquos where-abouts beyond whatrsquos necessary for determining proximity to infected individuals Localized data may also raise the same issues if accessed by law enforcement
Following Carpenter several courts have addressed the constitution-ality of novel location tracking In Massachusetts for instance a federal district court concluded that police use of a ldquopole camerardquo on a utility pole to investigate the movements of an individual constituted a search under the Fourth Amendment United States v Moore-Bush 381 FSupp3d 139 (D Mass 2019) The court reasoned that even in a public space an in-dividual still retains a reasonable expectation of privacy ldquoin the whole of their physical movementsrdquo Citing Carpenter and Jones the court stated that the governmentrsquos unrestrained power to collect data that reveal pri-vate aspects of identity is susceptible to abuse and gives police access to a category of information that is ldquootherwise unknowablerdquo Long-term monitoring of a personrsquos movements consequently violates that individ-ualrsquos expectation of privacy Notably the court emphasized the capability of the camera to create a searchable digital log of the photos taken for the eight-month period during which the camera was used
State courts have also weighed in on the issue The Massachusetts Supreme Judicial Court found that police access to real-time location data pinpointing an individualrsquos movement whether from a third party or a cell-site simulator infringes upon an individualrsquos reasonable expec-tation of privacy Commonwealth v Almonor 120 NE3d 1183 1195 (Mass 2019) The Washington Supreme Court for its part held that a cell phone ping used to locate the defendantrsquos vehicle in real time is a search under the Fourth Amendment requiring a warrant absent exigent circumstances State v Muhammad 428 P3d 1177 (2018) And the Colo-
90 Digital Contact Tracing for Pandemic Response
rado Court of Appeals held that police use of a video pole camera to con-tinuously surveil a defendantrsquos fenced-in backyard constitutes a search under the Fourth Amendment People v Tafoya 2019 BL 457321 Colo Ct App 17CA1243 (2019)
Application of Carpenter by lower courts to novel location-tracking tactics is still evolving and it is as yet unclear how the narrower interpre-tation of the third-party doctrine will continue to be expanded and ap-plied particularly in cases of short-term monitoring of massive amounts of location andor health data Moreover it is unclear whether Carpenter would apply to DCTT data collected by the government itself
Special Needs Doctrine
An argument in favor of the constitutionality of government DCTT programs is that the ldquospecial needsrdquo doctrine would apply Under this doctrine a warrantless search that would otherwise violate the Fourth Amendment might be permissible based on a special need relating to pub-lic health When the search is conducted for a nontraditional law enforce-ment purpose and circumstances make securing a warrant impracticable the Supreme Court has ruled that warrantless searches may be permissi-ble The special needs doctrine however is highly controversial because it is not a consistently applied Fourth Amendment exception so it is diffi-cult to predict when courts would authorize nontraditional surveillance Some factors considered by the court are (1) the balance between the intrusiveness of the government action and the anticipated public bene-fits (2) the existence of legislative authorization (3) judicial process or the ability of the subject individual to challenge the government action (4) the scope or breadth of government action and (5) the likelihood of the collected data being used in criminal proceedings The Supreme Court did note in Chandler v Miller 520 US 305 (1997) that a ldquorisk to public safety [that] is substantial and realrdquo may justify ldquoblanket suspicionless searches calibrated to the riskrdquo citing as examples the routine searches conducted at airports and entrances to some official buildings (Searches within the context of immigration are further analyzed below)
Immigration Enforcement
Exceptions apply to the constitutional requirement that a warrant ac-company an unreasonable search or seizure in the immigration context
Legal Considerations 91
For example an exception to the general warrant requirement is the bor-der search exception which allows government officials to search and seize without a warrant persons and property at the border or at the functional equivalent of a border See United States v Montoya de Her-nandez 473 US 531 (1985) United States v Flores-Montano 541 US 149 (2004) Federal regulation authorizes immigration officials to oper-ate within 100 miles of any US external boundary (See 8 CFR sect 2871 defining ldquoreasonable distancerdquo as ldquowithin 100 air miles from any external boundary of the United Statesrdquo) A functional equivalent of a border may include any airport where international flights may be received automo-bile checkpoints servicing international traffic and vessels in territorial waters Government officials however must still have ldquoreasonable suspi-cionrdquo of an immigration violation or a crime to search or seize persons or property
In the context of digital data Customs and Border Protection (CBP) officials may conduct either manual or forensic searches of electronic devices at the border or its functional equivalent A manual search is considered a routine search and may include accessing the phone and ldquobrowsingrdquo its contents If the electronic device is password protected individuals must provide information for unlocking the device Forensic searches on the other hand are nonroutine and involve a more invasive search of the electronic devicersquos contents Federal circuit courts are split on whether a CBP agent needs ldquoreasonable suspicionrdquo before conducting a forensic search of an electronic device But Supreme Court precedent clearly states that suspicionless searches are not unconstitutional when public safety is considered Skinner v Ry Labor Execsrsquo Assrsquon 489 US 602 (1989)
A recent CBP directive provides guidance and standard operating procedures regarding forensic searches of electronic devices CBP 3340-049A Border Search of Elec Devices (DHS 2018) The directive states that CBP officers may detain electronic devices or copies of the informa-tion contained within these devices for a reasonable period time not to exceed five days This directive raises the concern that travelers may be required to turn over contact tracing data stored on their phone to CBP officers Note that the directive has been challenged in federal court and is currently awaiting appeal Alasaad v Nielsen 419 FSupp3d 142 (D Mass 2019)
92 Digital Contact Tracing for Pandemic Response
Searches in Schools
Another exception to the general warrant requirement applies to searches by non-law-enforcement government officials in public schools (ie school officials) Within this context school officials have broad powers to conduct searches as long as those searches are reasonable Searches by individuals in private schools are not governed by the Fourth Amend-ment State regulation of searches in private schools varies (See US DOE 2009)
Related Federal Privacy Statutes
Outside the Fourth Amendment context certain laws provide protections against government collection of and access to personal data The USA Freedom Act of 2015 for example bans the governmentrsquos bulk collec-tion of internet metadata and telephonic records which was previously allowed under Section 215 of the USA Patriot Act The government must now identify with specificity the identity of a person account address or personal device when requesting records The law allows for the acqui-sition of data by two degrees of separationmdashor ldquohopsrdquomdashfrom targeted individuals If a centralized system in contact tracing is used it is unclear whether the government may need to resort to this provision since it would likely have consent from individuals to collect and use the data
The Privacy Act of 1974 also regulates the collection use and disclo-sure of personal data but applies only to federal agencies (and their con-tractors) not to state or local agencies 5 USC sect 552a The Act protects against disclosure of individually identifying ldquorecord[s]rdquo that are kept within a ldquosystem of recordsrdquo The Act limits disclosure of information ldquoexcept pursuant to a written request by or with prior written consent of the individual to whom the record pertainsrdquo Certain disclosures are ex-empt from the Actrsquos applicability Pertinent disclosure exceptions are for records required to be disclosed under the Freedom of Information Act (FOIA) or disclosures ldquoto a person pursuant to a showing of compelling circumstances affecting the health or safety of an individualrdquo A disclosure under FOIA however would not include information in ldquopersonnel and medical files and similar filesrdquo when disclosure ldquowould constitute a clearly unwarranted invasion of personal privacyrdquo FOIA Guide 2004 Edition
Legal Considerations 93
Exemption 6 If non-anonymized data are turned over to the federal or state governments it is important to consider whether PII would be pro-tected from disclosure under FOIA or state freedom of information laws
Consent
User consent is a cross-cutting issue for evaluating many of the laws and regulations governing personal information privacy discussed in the prior sections In general privacy laws can be justified on the grounds that an individual should have the option to control with various types and de-grees of limitation the collection use retention andor disclosure of in-formation pertaining to that individual by others As such many privacy laws start from the premise that absent an individualrsquos consent use or disclosure of that individualrsquos PII is impermissible except for certain enu-merated purposes deemed to outweigh the individualrsquos privacy interests
Consent like agreements in general can be manifest in different ways in specific circumstances In some cases an affirmative actionmdashsuch as a signaturemdashis needed to demonstrate consent In other cases inactionmdashsuch as declining to ldquounsubscriberdquo from receiving certain unsolicited emailsmdashconstitutes consent Where a law requires a ldquowrittenrdquo signature in all but a few contexts the signature may be executed electronically In the United States that means the ldquosignaturerdquo may consist of any of the following ldquoan electronic sound symbol or processrdquo so long as it is ldquoattached to or logically associated with a contract or other record and executed or adopted by a person with the intent to sign the recordrdquo Electronic Signatures in Global and National Commerce Act (E-SIGN) (15 USC 7006)
The scope of a consent depends on what was deemed to be under-stood by the consenting party That is least clear when the consent is in-ferred from inactivity even if terms stating the consequences of inactivity have been provided The scope of consent is most clear when the terms agreed to have been presented to or provided by the consenting party in a conspicuous documented manner and a record exists of those terms Courts uphold the validity of clickthrough agreements because users are deemed to review the terms to which they respond by clicking ldquoI agreerdquo
94 Digital Contact Tracing for Pandemic Response
But where terms are ambiguous or confusing buried in other text or presented obscurely the ldquoI agreerdquo action may not mean the user actually agreed to specific terms
Terms of Use and Privacy Policies for apps are often written in com-plicated or nuanced language with key points difficult to discern More-over they are generally hard to read on a mobile device Many users of mobile phone apps agree to such terms without even attempting to read or to understand them As such it is often questionable whether an app user has knowingly agreed to all the terms of those documents Presen-tation of terms in large typeface short sentences simple language and direct disclosures makes user consent more meaningful
For contact tracing apps that collect PII andor PHI consent will overcome the restrictions of many if not most privacy laws provided the consent is freely given reflects a full understanding of the terms for use collection and disclosure of the information and is confirmed by an af-firmative act such as a click that may be executed only upon a complete reading of Terms of Use and Privacy Policies Whether consent may be deemed ldquofreely givenrdquo in certain circumstances depends on contextual understandings of party relationships including the employer-employee and government-citizen relationships
Anti-discrimination and Individual Freedom Laws
Any measure taken to protect public health and safety must comply with the Constitution and civil rights laws such as the ADA that prohibit discrimination against persons in certain protected categories such as race gender religion or disability In addition certain implementations of DCTT could be challenged under a variety of individual freedom protections
Anti-discrimination Laws
In general it would be impermissible to use DCTT in a way that either targets or excludes people on the basis of their membership in one of these protected categories
When motivated by animus against a protected class as defined by law and not narrowly tailored to advance a compelling government inter-
Legal Considerations 95
est a discriminatory regulation would be considered unconstitutional un-der the Equal Protection Clause of the Fourteenth Amendment to the US Constitution See Jew Ho v Williamson 103 F10 CCND Cal (1900) (striking down a quarantine imposed by San Francisco in response to an outbreak of bubonic plague because it was racially motivated) see also Church of Lukumi Babalu Aye v Hialeah 508 US 520 (1993) (supposed public health measure unconstitutional because it targeted the practices of one religion)
The risk of unintentional yet illegal discrimination in using DCTT is a real possibility Recent studies of infection rates among the population have revealed a larger-than-proportional infection rate among certain minority communities such as Latinx African American and American Indian communities (NYC DOH 2020) Programs that target specific ra-cial ethnic tribal or religious groups may raise constitutional and other legal concerns
Religious Freedom Laws
The use of DCTT may also raise concerns about religious freedom For example there may be religious objections to restrictions on gathering for worship carrying a mobile phone or the use of imaging technology Under current Supreme Court precedent generally applicable laws that do not discriminate against religion on their face do not violate the Free Exercise Clause of the First Amendment even if those laws have an inci-dental effect on the exercise of religion Employment Div Dept of Hu-man Resources of Oregon v Smith 494 US 872 (1990) These laws need not be justified by compelling government interest (the ldquostrict scrutinyrdquo standard of review) the government need only show that they are ratio-nally related to a legitimate interest On the other hand laws that are not neutral and not of general applicability must be justified by compelling government interest and must be narrowly tailored to advance that in-terest if it burdens religious practicesmdasha very tough hurdle to overcome Church of Lukumi Babalu Aye v Hialeah 508 US 520 (1993)
This general approach however is disrupted in some contexts by statutes adopted to provide greater protection to religious freedom The federal Religious Freedom Restoration Act (RFRA) requires strict scru-tiny for federal actions that burden religion and many states have ad-opted ldquostate RFRAsrdquo that do the same for actions by state and local
96 Digital Contact Tracing for Pandemic Response
governments The Religious Land Use and Institutionalized Persons Act which extends similar protections to persons confined to an institution such as a prison jail or mental health facility may also be relevant 42 USC sect 2000cc
Under either standard of review courts will examine whether a gov-ernment action imposes a substantial burden on religious exercise if not no religious freedom violation has occurred Such a finding is unlikely for DCTT programs absent evidence that the government is using the digital information to take action against religious persons that is not necessary to avoid the spread of a serious disease Nevertheless legal challenges on religious freedom grounds cannot be ruled out
Legislative Recommendations
bull Congress should enact new legislation specifically tailored to fa-cilitate the use of DCTT as part of the public health response to COVID-19 while also protecting user privacy and ensuring data security
bull Congress should require DCTT developers to disclose to users in clear language the nature of the information that would be col-lected how it would be collected how it would be stored and for what purposes it may be used
bull While the rollout of DCTT should initially employ an opt-in au-thorization approach the feasibility acceptability and value of opt-out approaches should continue to be evaluated As such opt-out approaches to consent should not be precluded by legislation
bull Congress should prohibit the commercial use of data collected for COVID-19 response by DCTT
bull Congress should prohibit discrimination on the basis of data col-lected by DCTT
bull If Congress is unable to enact suitable legislation state legislatures should work toward enacting similar laws for their jurisdictions A ldquomodelrdquo state law should be rapidly developed to facilitate na-tionwide uniformity of legal requirements
97
Public Health
Characteristics That Make Data Useful to
Public Health for Reducing Disease Transmission
bull Technologies or apps with the goal of enhancing public health ca-pacity to identify cases and trace contacts in order to control the spread of SARS-CoV-2 should be designed to match functionality with that goal
bull Technologies or apps may produce some false negatives or false positives but they should be accurate enough that public health authorities feel confident that they support and donrsquot detract from contact tracing efforts
bull DCTT approaches for public health should be designed to facili-tate the following
deg Identifying contacts including those who may not be easily found otherwise
deg Finding and notifying contacts rapidly before they develop symptoms if infected
deg Analyzing the nature of contact to determine whether con-tact is high medium or low risk to support decisions about whether quarantine should be mandatory should be volun-tary or is not needed
deg Following up with cases and contacts so that public health can provide resources to support isolation and quarantine at home
Recommendations
FIVE
98 Digital Contact Tracing for Pandemic Response
bull Data collected through DCTT should be made available to public health professionals and to researchers in de-identified form to support population-level epidemiologic analysis
Ethics
Collecting Data to Inform Policy and Practice
bull Reviews of DCTT systems must be conducted in part by an inde-pendent intermediary that has established the publicrsquos trust
bull Those who authorize use of DCTT within a particular jurisdic-tion or institution should continuously and systematically monitor the technologyrsquos performance in that context This should include monitoring for effectiveness and benefit monitoring for harms and monitoring for the fair distribution of both benefits and harms They should also monitor evidence that is being generated in other contexts about their selected technological solution and about other competing technologies
bull Data should be available to users that would permit them to further investigate their personal risk with public health officials or other health workers to add a layer of protection against unnecessary quarantine
Public Trust and Public Attitudes
bull More research into public attitudes is needed In particular in-depth qualitative research should examine public attitudes about perceptions of trust in DCTT among different communities which features of DCTT influence trust and the extent to which people are willing to provide different types of data through DCTT to help their community
bull States and localities that are considering adopting DCTT should engage with the public to increase their understanding of the acceptability of DCTT design features and uses among diverse communities
Recommendations 99
Designing Flexible Technology to Maximize
Public Health Utility While Respecting Other Values
bull Technology companies should not alone control the terms condi-tions or capabilities of DCTT nor should they presume to know what may be acceptable to members of the public
bull A ldquovalues in designrdquo approach to development of DCTT should be adopted (Flanagan Howe and Nissenbaum 2008 Knobel and Bowker 2011) Robust public- and user-engagement activities should be pursued to identify and incorporate to the extent pos-sible a range of values into the design of the technology These values may include privacy but also autonomy efficiency equity or others Technology design should reflect an appropriate balance and prioritization of identified values
bull Technology design should not be static but should be capable of evolving depending upon local conditions new evidence and changing preferences and priorities
bull DCTT should be designed to have a base set of features that pro-tect privacy with layers of additional capabilities that users may choose to activate An initial default should be that user location data are not shared but users should be provided with easy mech-anisms and prompts to allow for opting-in to this capability with encouragement to the public if and as it is shown to be critical to achieving public health goals
Policy Positions to Advance Widespread
Use of Digital Contact Tracing Technologies
bull DCTT use should not be mandated at this time given uncertainty about potential harms and benefits Additional technology user and real-world testing is needed
bull Incentives can be a useful complement to encouragements how-ever any incentives for users to install and use DCTT must be equitable should not be coercive and should align with effective use of the technology
100 Digital Contact Tracing for Pandemic Response
bull Trusted leaders should be enlisted to communicate effectively with the public about DCTT and encourage its use should the technol-ogy demonstrate some potential The limits of knowledge regard-ing effectiveness should also be explained along with what will be done to improve technological capabilities as understanding evolves
Disclosure and AuthorizationConsent
bull A clear and concise module consisting of basic disclosure and vol-untary authorization should be developed to accompany DCTT This module should not take the form of ldquoclickwraprdquo terms of service or end-user agreements but rather provide only essential information necessary for an individual to make a decision More detailed disclosures (such as FAQs in plain language) should be made easily accessible to those who wish to learn more with no hidden surprises
bull An opt-in approach to authorization should be instituted to ac-company initial DCTT rollout The feasibility and value of opt-out approaches should continue to be evaluated informed by what is technologically possible what local assessments of benefits and harms of the technology reveal over time and our evolving un-derstanding of the degree to which an opt-out approach is likely to increase or decrease utilization among different populations Opt-out approaches should not be precluded
Promoting Equity and Fairness
bull A commitment to equity means a commitment to ensuring that the benefits and burdens of DCTT are distributed fairly Public engagement is an important tool for assessing impact and to rectify inequities
bull States localities and institutions that recommend widespread use of DCTT should provide technology (eg mobile phones Blue-tooth devices) and free data packages to those who desire but lack access to these devices
Recommendations 101
bull If there are lower rates of adoption of DCTT systems in some identifiable communities public health authorities should iden-tify ways to compensate For example directing more non-DCTT resources and efforts toward those communities to meet specific needs that are elsewhere being supported by technology
bull If maps are generated based on DCTT to provide the public with the locations that COV+ individuals have visited steps must be taken to minimize the stigma and potential financial losses that could result from being identified as a hotspot
Instituting Transparent Governance and Oversight
bull Digital surveillance oversight committees should be established expeditiously with diverse and qualified membership to provide ethical and regulatory review prior to and concurrent with wide-spread use of a DCTT system
bull Understandable and publicly accessible rules must guide the col-lection access control use storage and combination of data by government authorities public and private institutions and other parties such as public health researchers
bull Only those data that are necessary and relevant for the public health response to COVID-19 should be collected and used
bull Identifiable data should be kept only for the period of time needed for the public health response to COVID-19
bull Identifiable data collected as part of this response should not be shared with anyone other than the relevant public health authorities without additional specific informed consent of individual users
bull Before a government or institution adopts a digital contact tracing program they should state the conditions under which the digital contact tracing program will be terminated
bull Future use of DCTT to advance public health or other efforts (eg use in seasonal flu surveillance) would require independent justification DCTT designed for public health use should not be used by law or immigration enforcement
102 Digital Contact Tracing for Pandemic Response
bull The principles offered in this guidance document apply both during and following the COVID-19 pandemic
Legislative
bull Congress should enact new legislation specifically tailored to fa-cilitate the use of DCTT as part of the public health response to COVID-19 while also protecting user privacy and ensuring data security
bull Congress should require DCTT developers to disclose to users in clear language the nature of the information that would be col-lected how it would be collected how it would be stored and for what purposes it may be used
bull While the rollout of DCTT should initially employ an opt-in au-thorization approach the feasibility acceptability and value of opt-out approaches should continue to be evaluated As such opt-out approaches to consent should not be precluded by legislation
bull Congress should prohibit the commercial use of data collected for COVID-19 response by DCTT
bull Congress should prohibit discrimination on the basis of data col-lected by DCTT
bull If Congress is unable to enact suitable legislation state legislatures should work toward enacting similar laws for their jurisdictions A ldquomodelrdquo state law should be rapidly developed to facilitate na-tionwide uniformity of legal requirements
Resources
103
US Government Response
White House
The White House and CDC 16 April 2020 ldquoOpening Up America Againrdquo The White House and the Centers for Disease Control and Prevention Available at httpswwwwhitehousegovopeningamerica
Congress
A Bill to Protect the Privacy of Consumersrsquo Personal Health Information
Proximity Data Device Data and Geolocation Data during the Corona-
virus Public Health Crisis S3663 116th Cong (2019-2020) Available at httpswwwcongressgovbill116th-congresssenate-bill3663
A Bill to Protect the Privacy of Health Information during a National
Health Emergency S3749 116th Cong (2019-2020) Available at httpswwwcongressgovbill116th-congresssenate-bill3749
To Protect the Privacy of Health Information during a National Health
Emergency HR6866 116th Cong (2019-2020) Available at httpswwwcongressgovbill116th-congresshouse-bill6866
Congressional Research Service
Foster Michael 16 April 2020 ldquoCOVID-19 Digital Surveillance and Privacy Fourth Amendment Considerationsrdquo Legal Sidebar LSB10449 Congres-sional Research Service httpscrsreportscongressgovproductpdfLSBLSB10449
Mulligan Stephen P and Chris D Linebaugh 25 March 2019 ldquoData Pro-tection Law An Overviewrdquo R45631 Congressional Research Service httpscrsreportscongressgovproductpdfRR45631
104 Resources
US Department of Health and Human Services (HHS) Centers for Disease Control and Prevention (CDC)
CDC 6 May 2020 ldquoInterim Guidance for Businesses and Employersrdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention Available at httpswwwcdcgovcoronavirus2019-ncov communityguidance-business-responsehtml
CDC 3 May 2020 ldquoDiscontinuation of Isolation for Persons with COVID-19 Not in Healthcare Settingsrdquo Interim Guidance Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention httpswwwcdc govcoronavirus2019-ncovhcpdisposition-in-home-patientshtml
CDC 30 April 2020 ldquoContact Tracingrdquo Get and Keep America Open Sup-porting States Tribes Localities and Territories Centers for Disease Control and Prevention httpswwwcdcgovcoronavirus2019-ncovphpopen-americacontact-tracinghtml
CDC 29 April 2020 ldquoInterim Guidelines for Collecting Handling and Testing Clinical Specimens from Persons for Coronavirus Disease 2019 (COVID-19)rdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention Available at httpswwwcdcgovcorona-virus2019-ncovlabguidelines-clinical-specimenshtml
CDC 29 April 2020 ldquoContact Tracing Part of a Multipronged Approach to Fight the COVID-19 Pandemicrdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention httpswwwcdcgov coronavirus2019-ncovphpprinciples-contact-tracinghtml
CDC 28 April 2020 ldquoPreliminary Criteria for the Evaluation of Digital Contact Tracing Tools for COVID-19rdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention Available at httpswwwcdc govcoronavirus2019-ncovdownloadsphpprelim-eval-criteria-digital- contact-tracingpdf
CDC 20 April 2020 ldquoDigital Contact Tracing Tools for COVID-19rdquo Corona- virus Disease 2019 (COVID-19) Centers for Disease Control and Preven-tion Available at httpswwwcdcgovcoronavirus2019-ncovdownloadsdigital-contact-tracingpdf
CDC 6 April 2020 ldquoOperational Considerations for the Identification of Healthcare Workers and Inpatients with Suspected COVID-19 in Non-US Healthcare Settingsrdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention Available at httpswwwcdcgovcoron virus2019-ncovhcpnon-us-settingsguidance-identify-hcw-patientshtml
HHS 2 April 2020 ldquoNotification of Enforcement Discretion under HIPAA to Allow Uses and Disclosures of Protected Health Information by Business
Resources 105
Associates for Public Health and Health Oversight Activities in Response to COVID-19rdquo 45 CFR Parts 160 and 164 Available at httpswwwhhsgovsitesdefaultfilesnotification-enforcement-discretion-hipaapdf
HHS 30 March 2020 Notification of Enforcement Discretion for Telehealth Remote Communications during the COVID-19 Nationwide Public Health Emergency US Department of Health and Human Services Office of Civil Rights Available at httpswwwhhsgovhipaafor-professionalsspecial- topicsemergency-preparednessnotification-enforcement-discretion- telehealthindexhtml
CDC 24 February 2020 ldquoLegal Authorities for Isolation and Quarantinerdquo Centers for Disease Control and Prevention Available at httpswww cdcgovquarantineaboutlawsregulationsquarantineisolationhtml
HHS 26 November 2019 ldquoPublic Health Emergency Declarationrdquo Public Health Emergency US Department of Health and Human Services Office of the Assistant Secretary for Preparedness and Response Available at httpswwwphegovPreparednesslegalPagesphedeclarationaspx
CDC September 2018 ldquoPublic Health Surveillance Preparing for the Futurerdquo Centers for Disease Control and Prevention Available at httpswww cdcgovsurveillancepdfsSurveillance-Series-Booklethpdf
CDC 2005 ldquoVI Plan for Surveillance of Contacts of SARS Case Supplement B SARS Surveillancerdquo Centers for Disease Control and Prevention Available at httpswwwcdcgovsarsguidanceb-surveillancecontactshtml
National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research 1979 ldquoThe Belmont Report Ethical Principles and Guidelines for the Protection of Human Subjects of Researchrdquo US Depart-ment of Health and Human Services Office of Human Research Protec-tions httpswwwhhsgovohrpregulations-and-policybelmont-reportread-the-belmont-reportindexhtml
US Equal Employment Opportunity Commission (EEOC)
EEOC 7 May 2020 ldquoWhat You Should Know About COVID-19 and the ADA the Rehabilitation Act and Other EEO Lawsrdquo US Equal Employ-ment Opportunity Commission Available at httpswwweeocgovwyskwhat-you-should-know-about-covid-19-and-ada-rehabilitation-act-and-other-eeo-laws
EEOC 21 March 2020 ldquoPandemic Preparedness in the Workplace and the Americans with Disabilities Actrdquo US Equal Employment Oppor-tunity Commission Available at httpswwweeocgovlawsguidancepandemic-preparedness-workplace-and-americans-disabilities-act
106 Resources
Federal Trade Commission (FTC)
FTC 2020 ldquoPrivacy amp Data Security Update 2019rdquo Federal Trade Commis-sion httpswwwftcgovsystemfilesdocumentsreportsprivacy-data- security-update-20192019-privacy-data-security-report-508pdf
FTC 2012 ldquoProtecting Consumer Privacy in an Era of Rapid Changerdquo FTC Report Recommendations for Businesses and Policymakers Fed-eral Trade Commission httpswwwftcgovsitesdefaultfilesdocumentsreportsfederal-trade-commission-report-protecting-consumer-privacy-era- rapid-change-recommendations120326privacyreportpdf
Occupational Safety and Health Administration (OSHA)
OSHA April 2020 ldquoGuidance on Preparing Workplaces for COVID-19rdquo OSHA 3990-03 2020 Department of Labor Occupational Safety and Health Act of 1970 Available at httpswwwoshagovPublicationsOSHA3990pdf
Customs and Border Protection (CBP)
CBP 4 January 2018 ldquoBorder Search of Electronic Mediardquo US Customs and Border Protection CDP DIRECTIVE NO 3340-049A Available at httpswwwcbpgovsitesdefaultfilesassetsdocuments2018-JanCBP-Directive-3340-049A-Border-Search-of-Electronic-Media-Compliantpdf
Other Governmental and Nongovernmental Organizations
Contact Tracing Surveillance PlansMethods
Simmons-Duffin Selena 7 May 2020 ldquoStates Nearly Doubled Plans for Contact Tracers since NPR Surveyed Them 10 Days Agordquo NPR Available at httpswwwnprorgsectionshealth-shots20200428846736937we-asked-all-50-states-about-their-contact-tracing-capacity-heres-what- we-learned
Resolve to Save Lives 2020 ldquoCOVID-19 Contact Tracing Playbookrdquo Vital Strategies Available at httpscontacttracingplaybookresolvetosavelives org
Simpson Erin and Adam Conner 22 April 2020 ldquoDigital Contact Tracing to Contain the Coronavirusrdquo Center for American Progress Avail-able at httpswwwamericanprogressorgissuestechnology-policynews20200422483521digital-contact-tracing-contain-coronavirus
Watson Crystal Anita Cicero James Blumenstock and Michael Fraser 10 April 2020 ldquoA National Plan to Enable Comprehensive COVID-19 Case Finding and Contact Tracing in the USrdquo Johns Hopkins Bloomberg School of Pub-
Resources 107
lic Health Center for Health Security and the Association of State and Ter-ritorial Health Officials Available at httpswwwcenterforhealthsecurity orgour-workpubs_archivepubs-pdfs2020200410-national-plan-to- contact-tracingpdf
Africa CDC 9 April 2020 ldquoGuidance on Contact Tracing for COVID-19 Pandemicrdquo Manuals Guidelines amp Frameworks African Union Africa CDC Available at httpsafricacdcorgdownloadguidance-on-contact- tracing-for-covid-19-pandemic
PIH 4 April 2020 ldquoPart I Testing Contact Tracing and Community Manage-ment of COVID-19rdquo PIH Guide | COVID-19 STOP COVID Partners in Health Available at httpswwwpihorgsitesdefaultfiles2020-04PIH_Guide_COVID_Part_I_Testing_Tracing_Community_Managment_4_4pdf
Hellewell Joel Sam Abbott Amy Gimma et al 28 February 2020 ldquoFea- sibility of Controlling COVID-19 Outbreaks by Isolation of Cases and Contactsrdquo The Lancet Global Health 8 e488ndash96 Available at https doiorg101016S2214-109X(20)30074-7
Contact Tracing EthicsPrinciplesGuidance
Center for Democracy amp Technology (CDT)ndashhttpscdtorginsightsCDT 30 April 2020 ldquoStatement of the Center for Democracy amp Technol-
ogy Regarding Use of Data to Fight COVID-19rdquo Center for Democ-racy amp Technology httpscdtorgwp-contentuploads202004 CDT-Statement-Regarding-Use-of-Data-to-Fight-COVID-19pdf
Shetty Ridhi 23 April 2020 ldquoData Use in the Fight against COVID-19 Should Treat People Equitably Not Exacerbate Long-Standing Dispar-itiesrdquo Center for Democracy amp Technology httpscdtorginsightsdata-use-in-the-fight-against-covid-19-should-treat-people-equitably-not-exacerbate-long-standing-disparities
Greenwood Dazza Gregory Nadeau Pagona Tsormpatzoudi Bryan Wilson Jeffrey Saviano and Alex ldquoSandyrdquo Pentland 30 April 2020 ldquoCOVID-19 Contact Tracing Privacy Principlesrdquo MIT Computational Law Report Available at httpslawmitedupubcovid19contacttracingprivacy- principlesrelease7
Editorial Board 29 April 2020 ldquoShow Evidence That Apps for COVID-19 Contact-Tracing Are Secure and Effectiverdquo Nature 580 (7805) 563ndash563 httpsdoiorg101038d41586-020-01264-1
ldquoJoint Statement on Contact Tracingrdquo 19 April 2020 Available at httpscryptobriefingcomwp-contentuploads202004Joint-State-ment-from-Researcherspdf
Hinch Robert Will Probert Anel Nurtay Michelle Kendall Chris Wymant
108 Resources
Matthew Hall Katrina Lythgoe et al 16 April 2020 ldquoEffective Configura-tions of a Digital Contact Tracing App A Report to NHSXrdquo Available at httpsgithubcomBDI-pathogenscovid-19_instant_tracingblobmasterReportndashEffective Configurations of a Digital Contact Tracing Apppdf
Kahn Gilmor Daniel 16 April 2020 ldquoPrinciples for Technology-Assisted Contact-Tracingrdquo White Paper American Civil Liberties Union httpswwwacluorgreportaclu-white-paper-principles-technology-assisted- contact-tracing
Carroll Anna and Samantha Stroman 16 April 2020 ldquoFind My Friends in a Pandemic The Future of Contact Tracing in Americardquo CSIS Commis-sion on Strengthening Americarsquos Health Security April 16 2020 httpshealthsecuritycsisorgarticlesfind-my-friends-in-a-pandemic-the-future- of-contact-tracing-in-america
Electronic Privacy Information Center Testimony to Congress 15 April 2020 ldquoEPIC to Congress Establish Privacy Safeguards for Digital Contact Tracingrdquo httpsepicorgtestimonycongressEPIC-HEC-Contact-Tracing-Apr2020pdf
Crocker Andrew Kurt Opsahl and Bennett Cyphers 10 April 2020 ldquoThe Chal-lenge of Proximity Apps for COVID-19 Contact Tracingrdquo Electronic Fron-tier Foundation April 10 2020 httpswwwefforgdeeplinks202004challenge-proximity-apps-covid-19-contact-tracing
Hart Vi Divya Siddarth Bethan Cantrell Lila Tretikov Peter Eckersley John Langford Scott Leibrand et al 3 April 2020 ldquoOutpacing the Virus Digital Response to Containing the Spread of COVID-19 While Mitigat-ing Privacy Risksrdquo Whitepaper 5 COVID-19 Rapid Response Impact Initiative Edmond J Safra Center for Ethics httpsdrivegooglecomfiled1vIN2AX-DDNW-S0aHq8xs0RJ2jkR_CckXview
Raskar Ramesh Isabel Schunemann Rachel Barbar Kristen Vilcans Jim Gray Praneeth Vepakomma Suraj Kapa Andrea Nuzzo Rajiv Gupta et al 19 March 2020 ldquoApps Gone Rogue Maintaining Personal Privacy in an Epidemicrdquo White Paper Private Kit MIT httpsarxivorgpdf2003 08567pdf
General (not COVID-19 specific) Statements of Principles Legal Frameworks Other Information
Nuffield Council on Bioethics 2020 ldquoGuide to the Ethics of Surveillance and Quarantine for Novel Coronavirusrdquo httpswwwnuffieldbioethicsorgassetspdfsGuide-to-the-ethics-of-surveillance-and-quarantine-for-novel-coronaviruspdf
Resources 109
Schwartz Adam 2020 ldquoHow EFF Evaluates Government Demands for New Surveillance Powersrdquo Electronic Frontier Foundation April 3 2020 httpswwwefforgdeeplinks202004how-eff-evaluates-government-demands- new-surveillance-powers
WHO 2 April 2020 ldquoCoronavirus Disease 2019 (COVID-19)rdquo Situation Report 73 World Health Organization httpswwwwhointdocsdefault-sourcecoronavirusesituation-reports20200402-sitrep-73-covid-19pdf
Nuffield Council on Bioethics 17 March 2020 ldquoEthical Considerations in Responding to the COVID-19 Pandemicrdquo Rapid Policy Briefing Nuffield Council on Bioethics httpswwwnuffieldbioethicsorgassetspdfs Ethical-considerations-in-responding-to-the-COVID-19-pandemicpdf
Nuffield Council on Bioethics 28 January 2020 ldquoResearch in Global Health Emergencies Ethical Issuesrdquo httpswwwnuffieldbioethicsorgassetspdfsRGHE_full_report1pdf
Vota Wayan 4 December 2019 ldquoEvery African Countryrsquos National eHealth Strategy or Digital Health Policyrdquo ICT Works httpswwwictworksorgafrican-national-ehealth-strategy-policy
ENISA 3 December 2019 ldquoPseudonymisation Techniques and Best Practicesrdquo European Union Agency for Cybersecurity httpswwwenisaeuropaeupublicationspseudonymisation-techniques-and-best-practices
ldquoMobile Cellular Subscriptions (per 100 People)rdquo 2018 World Telecommuni-cationICT Development Report International Telecommunication Union httpsdataworldbankorgindicatorITCELSETSP2
Levinson-Waldman Rachel 2018 ldquoCellphones Law Enforcement and the Right to Privacyrdquo Brennan Center for Justice New York University School of Law httpswwwbrennancenterorgsitesdefaultfiles2019-08Report_Cell_Surveillance_Privacypdf
Code of Federal Regulations Part 46ndashProtection of Human Subjects (45 CFR 46 US Department of Health and Human Services)
General Data Protection Regulation (REGULATION (EU) 2016679 (GDPR) European Union)
European Data Protection Supervisor ldquoNecessity and Proportionalityrdquo httpsedpseuropaeudata-protectionour-worksubjectsnecessity- proportionality_en
ISOIEC 38505-12017 Information technologymdashGovernance of ITmdashGover-nance of datamdashPart 1 Application of ISOIEC 38500 to the governance of data (2017 International Organization for Standardization (ISO))
WHO 2016 ldquoWHO Guidance for Managing Ethical Issues in Infectious Dis-ease Outbreaksrdquo World Health Organization httpswwwwhointethicspublicationsinfectious-disease-outbreaksen
110 Resources
CIOMS 2016 ldquoInternational Ethical Guidelines for Health-Related Research Involving HumansndashGuideline 22 Use of Data Obtained from the Online Environment and Digital Tools in Health Related Researchrdquo Council for International Organizations of Medical Sciences httpsciomschwp- contentuploads201701WEB-CIOMS-EthicalGuidelinespdf
Principles for Digital Development 2016 ldquoPrinciplesrdquo httpsdigitalprinciplesorgprinciples
GA4GH 2014 ldquoFramework for Responsible Sharing of Genomic and Health-Related Datardquo Global Alliance for Genomics amp Health 9 Decem-ber 2014 httpswwwga4ghorggenomic-data-toolkitregulatory-ethics- toolkitframework-for-responsible-sharing-of-genomic-and-health-related- data
WHO and PATH 2013 ldquoPlanning an Information Systems Project A Toolkit for Public Health Managersrdquo World Health Organization amp PATH httpspathazureedgenetmediadocumentsTS_opt_ict_toolkitpdf
ASTHO 2012 ldquoPublic Health Collection Use Sharing and Protection of Informationrdquo Issue Brief ASTHO Legal Preparedness Series Association of State and Territorial Health Officials httpswwwasthoorgProgramsPreparednessPublic-Health-Emergency-LawPublic-Health-and-Informat- ion-Sharing-ToolkitCollection-Use-Sharing-and-Protection-Issue-Brief
Cavoukian Ann 2010 ldquoPrivacy by Design The 7 Foundational Principlesrdquo Ontario Canada Information and Privacy Commissioner of Canada httpsiapporgmediapdfresource_centerpbd_implement_7found_ principlespdf
PHLS 2002 ldquoPrinciples of the Ethical Practice of Public Healthrdquo Public Health Leadership Society httpswwwaphaorg-mediafilespdfmembergroupsethicsethics_brochureashx
National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research 1979 ldquoThe Belmont Report Ethical Principles and Guidelines for the Protection of Human Subjects of Researchrdquo US Dept of Health amp Human Services Office of Human Research Protections httpswwwhhsgovohrpregulations-and-policybelmont-reportread-the-bel mont-reportindexhtml
Reopening ProposalsPlans (Broad)
United Kingdom Cabinet Office 2020 ldquoOur Plan to Rebuild The UK Govern-mentrsquos COVID-19 Recovery Strategyrdquo 12 May 2020 httpswwwgovukgovernmentpublicationsour-plan-to-rebuild-the-uk-governments-covid- 19-recovery-strategyour-plan-to-rebuild-the-uk-governments-covid-19- recovery-strategyfourteen-supporting-programmes
Resources 111
Shannon Joel Lorenzo Reyes and Doyle Rice 2020 ldquoAre Lockdowns Being Relaxed in My State Herersquos How America Is Reopening amid the Coronavirus Pandemicrdquo USA TODAY May 21 2020 httpswww usatodaycomstorynewshealth20200419coronavirus-lockdown- reopening-states-us-texas-florida5155269002
Romer Paul 2020 ldquoRoadmap to Responsibly Reopen Americardquo 23 April 2020 httpsroadmappaulromernetpaulromer-roadmap-reportpdf
Governor Larry Hogan 24 April 2020 ldquoMaryland Strong Roadmap to Recoveryrdquo httpsgovernormarylandgovwp-contentuploads202004MD_Strongpdf
NGA and ASTHO 21 April 2020 ldquoRoadmap to Recovery A Public Health Guide for Governorsrdquo National Governors Association and American State and Territorial Health Officials httpswwwngaorgwp-contentuploads202004NGA-Reportpdf
Allen Danielle Julius Krein Ganesh Sitaraman and E Glen Weyl 2020 ldquoNational Covid-19 Testing Action Planrdquo The Rockefeller Foundation httpswwwrockefellerfoundationorgwp-contentuploads202004TheRockefellerFoundation_WhitePaper_Covid19_4_22_2020pdf
Edmond J Safra Center for Ethics Harvard UniverityndashWhite PapersAllen Danielle Sharon Block Joshua Cohen et al 20 April 2020 ldquoRoad-
map to Pandemic Resiliencerdquo httpsethicsharvardedufilescenter- for-ethicsfilesroadmaptopandemicresilience_updated_42020_0pdf
Allen Danielle Lucas Stanczyk Rajiv Sethi Glen Weyl 25 March 2020 ldquoWhen Can We Go Outrdquo httpsdrivegooglecomfiled1gf21eYeNWwrR9OO5nzxn1jlv-RTmHkt0view
Mulheirn Ian Sam Alvis Lizzie Insall James Browne Christina Palmou 20 April 2020 ldquoA Sustainable Exit Strategy Managing Uncertainty Minimis-ing Harmrdquo Tony Blair Institute for Global Change httpsinstituteglobalsitesdefaultfilesinline-filesA20Sustainable20Exit20Strategy2C 20Managing20Uncertainty20Minimising20Harmpdf
Emanuel Zeke Neera Tanden Adam Conner Erin Simpson Nicole Rap-fogel and Maura Calsyn 2020 ldquoA National and State Plan to End the Coronavirus Crisisrdquo Center for American Progress April 3 2020 httpswwwamericanprogressorgissueshealthcarenews20200403482613national-state-plan-end-coronavirus-crisis
Gottlieb Scott Caitlin Rivers Mark McClellan Lauren Silvis and Crystal Watson 2020 ldquoNational Coronavirus Response A Road Map to Reopen-ingrdquo American Enterprise Institute httpswwwaeiorgresearch-productsreportnational-coronavirus-response-a-road-map-to-reopening
112 Resources
Digital Contact Tracing Experiences from Other Countries
Multiple
Woodhams Samuel 12 May 2020 ldquoCOVID-19 Digital Rights Trackerrdquo Top10VPN httpswwwtop10vpncomresearchinvestigationscovid-19- digital-rights-tracker
Fahim Kareem Min Joo Kim and Steve Hendrix 2 May 2020 ldquoCellphone Monitoring Is Spreading with the Coronavirus So Is an Uneasy Tolerance of Surveillancerdquo Washington Post (Washington DC) httpswww washingtonpostcomworldcellphone-monitoring-is-spreading-with-the- coronavirus-so-is-an-uneasy-tolerance-of-surveillance20200502 56f14466-7b55-11ea-a311-adb1344719a9_storyhtml
Ikram Umar Christer Mjaringset MD Anne-Marie Boxall Mylaine Breton Ines Gravey Holly Krelle Veacuteronique Raimond and Reginald D Williams II 30 April 2020 ldquoWhat Can the US Learn from Innovative Strategies Used in Other Countries to Respond to COVID-19rdquo The Commonwealth Fund httpswwwcommonwealthfundorgblog2020what-can-us-learn-innovative-strategies-used-other-countries-respond-covid-19
Jens-Henrik Jeppesen and Pasquale Esposito 29 April 2020 ldquoCOVID-19 European Data Collection and Contact Tracing Measuresrdquo Center for Democracy amp Technology httpscdtorginsightscovid-19-european- data-collection-and-contact-tracing-measures
Africa CDC 9 April 2020 ldquoGuidance on Contact Tracing for COVID-19 Pan-demicrdquo Manuals Guidelines amp Frameworks African Union Africa CDC httpsafricacdcorgdownloadguidance-on-contact-tracing-for-covid-19- pandemic
Heneghan Carl Jon Brassey and Tom Jefferson 6 April 2020 ldquoCOVID-19 What Proportion Are Asymptomaticrdquo Centre for Evidence-Based Medi-cine httpswwwcebmnetcovid-19covid-19-what-proportion-are- asymptomatic
Kharpal Arjun 30 March 2020 ldquoUse of Surveillance to Fight Coronavirus Raises Concerns about Government Power after Pandemic Endsrdquo CNBC httpswwwcnbccom20200327coronavirus-surveillance-used-by- governments-to-fight-pandemic-privacy-concernshtml
Canada
ldquoCommissioner Publishes Framework to Assess Privacy-Impactful Initiatives in Response to COVID19rdquo Office of the Privacy Commissioner of Canada 17 April 2020 httpswwwprivgccaenopc-newsnews-and-announcements 2020an_200417
Resources 113
China
Kraemer Moritz U G Chia-Hung Yang Bernardo Gutierrez Chieh-Hsi Wu Brennan Klein David M Pigott Open COVID-19 Data Working Group Louis du Plessis Nuno R Faria Ruoran Li William P Hanage John S Brownstein Maylis Layan Alessandro Vespignani Huaiyu Tian Chris-topher Dye Oliver G Pybus Samuel V Scarpino ldquoThe Effect of Human Mobility and Control Measures on the COVID-19 Epidemic in Chinardquo Science 368(6490) 493-497 DOI 101126scienceabb4218
Bi Qifang Yongsheng Wu Shujiang Mei Chenfei Ye Xuan Zou Zhen Zhang Xiaojian Liu Lan Wei Shaun A Truelove Tong Zhang Wei Gao Cong Cheng Xiujuan Tang Xiaoliang Wu Yu Wu Binbin Sun Suli Huang Yu Sun Juncen Zhang Ting Ma Justin Lessler and Teijian Feng ldquoEpidemiol-ogy and Transmission of COVID-19 in 391 Cases and 1286 of Their Close Contacts in Shenzhen China A Retrospective Cohort Studyrdquo The Lancet April 27 2020 DOIhttpsdoiorg101016S1473-3099(20)30287-5
Sun Kaiyuan and Ceacutecile Viboud ldquoImpact of contact tracing on SARS-CoV-2 transmissionrdquo The Lancet April 27 2020 DOIhttpsdoiorg101016S1473-3099(20)30357-1
Ferretti Luca Chris Wymant Michelle Kendall Lele Zhao Anel Nurtay Lucie Abeler-Doumlrner Michael Parker David Bonsall and Christophe Fraser 2020 ldquoQuantifying SARS-CoV-2 Transmission Suggests Epidemic Control with Digital Contact Tracingrdquo Science 368 (6491) httpsdoiorg101126scienceabb6936
Gan Nectar and David Culver 2020 ldquoChina Is Fighting the Coronavirus with a Digital QR code Herersquos How It Worksrdquo CNN Business April 16 2020 httpswwwcnncom20200415asiachina-coronavirus-qr-code-intl-hnkindexhtml
Mozur Paul Raymond Zhong and Aaron Krolik 2020 ldquoIn Coronavirus Fight China Gives Citizens a Color Code with Red Flagsrdquo New York Times (New York NY) March 1 2020 httpswwwnytimescom20200301businesschina-coronavirus-surveillancehtml
Germany
Schwartz Matthew S 2020 ldquoGermany Backs Away from Compiling Coro-navirus Contacts in a Central Databaserdquo NPR April 27 2020 https wwwnprorgsectionscoronavirus-live-updates20200427846046185 germany-backs-away-from-compiling-coronavirus-contacts-in-a-central- databaseutm_medium=RSSamputm_campaign=news
Busvine Douglas and Andreas Rinke 2020 ldquoGermany Flips to Apple-Google
114 Resources
Approach on Smartphone Contact Tracingrdquo Reuters April 26 2020 httpswwwreuterscomarticleus-health-coronavirus-europe-techgermany-flips-on-smartphone-contact-tracing-backs-apple-and-google-idUSKCN22807J
Hong Kong
ldquo lsquoStayHomeSafersquo Mobile App User Guiderdquo The Government of the Hong Kong
Special Administrative Region May 20 2020httpswwwcoronavirusgov hkengstay-home-safehtml
India
Dixit Pranav 2020 ldquoIndiarsquos Contact Tracing App Is All But Mandatory So This Programmer Hacked It So That He Always Appears Saferdquo BuzzFeed News May 12 2020 httpswwwbuzzfeednewscomarticlepranavdixitindia-aarogya-setu-hacked
Greenberg Andy 2020 ldquoIndiarsquos Covid-19 Contact Tracing App Could Leak Patient Locationsrdquo WIRED May 6 2020 httpswwwwiredcomstoryindia-covid-19-contract-tracing-app-patient-location-privacy
Alderson Elliot ldquoAarogya Setu The Story of a Failurerdquo Medium May 6 2020 httpsmediumcomfs0c131yaarogya-setu-the-story-of-a-failure- 3a190a18e34
OrsquoNeill Patrick Howell ldquoIndia Is Forcing People to Use Its COVID App Unlike Any Other Democracyrdquo MIT Technology Review May 6 2020 httpswwwtechnologyreviewcom202005071001360india-aarogya- setu-covid-app-mandatory
Israel
Hendrix Steve and Ruth Eglash 2020 ldquoIsrael Is Using Cellphone Surveil-lance to Warn Citizens You May Already Be Infectedrdquo Washington Post (Washington DC) March 19 2020 httpswwwwashingtonpostcomworldmiddle_eastisrael-is-using-cellphone-surveillance-to-warn-citizens-you-may-already-be-infected2020031968267294-69e7-11ea-b199-3a9799c54512_storyhtml
Singapore
ldquoTraceTogether Safer Togetherrdquo A Singapore Government Agency Website accessed May 21 2020 httpswwwtracetogethergovsg
Resources 115
South Korea
Korean Ministry of Health and Welfare ldquoConfirmed Patient Movement Path Websiterdquo Central Accident Remediation Headquarters accessed May 21 2020 httpncovmohwgokrbdBoardList_RealdobrdId=1ampbrdGu-bun=12ampncvContSeq=ampcontSeq=ampboard_id=ampgubun=
Kim Max S 2020 ldquoSeoulrsquos Radical Experiment in Digital Contact Tracingrdquo The New Yorker April 17 2020 httpswwwnewyorkercomnewsnews-deskseouls-radical-experiment-in-digital-contact-tracing
Watson Ivan and Sophie Jeong 2020 ldquoCoronavirus Mobile Apps Are Surging in Popularity in South Koreardquo CNN Business February 28 2020 httpseditioncnncom20200228techkorea-coronavirus-tracking-appsindexhtml
United Kingdom
Hern Alex and Kate Proctor 2020 ldquoUK May Ditch NHS Contact-Tracing App for Apple and Google Modelrdquo The Guardian May 7 2020 httpswwwtheguardiancomtechnology2020may07uk-may-ditch-nhs-contact- tracing-app-for-apple-and-google-model
Lovejoy Ben 2020 ldquoHands-on with UKrsquos NHS Contact Tracing App as the Test Goes Liverdquo 9-5 Mac Blog May 7 2020 https9to5maccom2020 0507nhs-contact-tracing
ldquoCoronavirus Test Track and Trace Plan Launched on Isle of Wightrdquo Depart-ment of Health and Social Care Press Release May 4 2020 httpswww govukgovernmentnewscoronavirus-test-track-and-trace-plan-launched- on-isle-of-wight
Leprince-Ringuet Daphne 2020 ldquoContact-Tracing Apps Why the NHS Said No to Apple and Googlersquos Planrdquo ZDNet April 28 2020 httpswww zdnetcomarticlecontact-tracing-apps-why-the-nhs-said-no-to-apple- and-googles-plan
Specific Digital ProductsApps
COVID-19
Vota Wayan 2020 Additional Proposed Coronavirus Solutions Google Doc-ument httpsdocsgooglecomspreadsheetsd15hkhdtGNzx7oHkO8Y 2MOiY83JsHjqxL4MhMGvlA_J6Ieditgid=579623365
Apple and Google nd ldquoPrivacy-Preserving Contact Tracingrdquo Apple Accessed May 19 2020 httpswwwapplecomcovid19contacttracing
116 Resources
Woodhams Samuel 12 May 2020 ldquoCOVID-19 Digital Rights Trackerrdquo Top10VPN httpswwwtop10vpncomresearchinvestigationscovid-19-digital-rights-tracker
Starobinski David and Johannes Becker 30 April 2020 ldquoHow Apple and Goo-gle Will Let Your Phone Warn You If Yoursquove Been Exposed to the Corona-virusrdquo The Conversation Accessed May 21 2020 httptheconversation comhow-apple-and-google-will-let-your-phone-warn-you-if-youve-been-exposed-to-the-coronavirus-136597
Morrison Sara 2020 ldquoThe Apple-Google Contact Tracing Tool Gets a Beta Release and a New Risk Level Featurerdquo Vox April 24 2020 httpswwwvoxcomrecode202042421234420apple-google-contact-tracing- exposure-notification-update
Commonwealth Centre for Digital Health 9 April 2020 ldquo[Webinar] CWCDH Digital Response to COVID-19rdquo 5040 httpswwwyoutubecomwatchv=ZyE_KRWLtC8ampfeature=youtube
Johns Hopkins Medicine 23 April 2020 ldquoJohns Hopkins Medicine Remote Monitoring Program for Faculty Staff and Providers Exposed to COVID-19rdquo Johns Hopkins Medicine Occupational Health Services httpswww hopkinsmedicineorghsecovid19_emocha
MIT nd ldquoPrivate Kit Safe Paths Privacy-by-Design Covid19 Solutions Using GPS+Bluetooth for Citizens and Public Health Officialsrdquo Safepaths Accessed May 20 2020 httpssafepathsmitedu
Related Disease Detection Projects
ldquoCOVID Controlrdquo Google Play accessed May 21 2020 httpsplaygoogle comstoreappsdetailsid=jhueduJohnsHopkinsCOVIDControl
ldquoCOVID Symptom Trackerrdquo Created by Massachusetts General Hospital the Harvard TH Chan School of Public Health Kingrsquos College London and Stanford University School of Medicine accessed May 21 2020 httpscovidjoinzoecomus
ldquoSee How Your Community Is Moving around Differently Due to COVID-19rdquo Google COVID-19 Mobility Reports accessed May 21 2020 httpswww googlecomcovid19mobility
Drew David A Long H Nguyen Claire J Steves Cristina Menni Maxim Freydin Thomas Varsavsky Carole H Sudre M Jorge Cardoso Sebastien Ourselin Jonathan Wolf Tim D Spector Andrew T Chan and COPE Consortium 2020 ldquoRapid Implementation of Mobile Technology for Real-Time Epidemiology of COVID-19rdquo Science 05 May 2020 DOI 101126scienceabc0473
Resources 117
Tress Luke 2020 ldquoMaccabi Medial EarlySign Develop Algorithm to Identify High-Risk COVID-19 Casesrdquo Times of Israel April 22 2020 httpswwwtimesofisraelcommaccabi-medial-earlysign-develop-algo rithm-to-identify-high-risk-covid-19-cases
ldquoPCR Diagnostic Testing for SARS-CoV-2rdquo Center for Health Security Johns Hopkins Bloomberg School of Public Health last modified April 17 2020 httpswwwcenterforhealthsecurityorgresourcesCOVID-19COVID-19-fact-sheets200130-nCoV-diagnostics-factsheetpdf
Radin Jennifer M Nathan E Wineinger Eric J Topol and Steven R Steinhubl 2020 ldquoHarnessing Wearable Device Data to Improve State-Level Real-Time Surveillance of Influenza-Like Illness in the USA A Population-Based Studyrdquo The Lancet January 16 2020 DOIhttpsdoiorg101016S2589-7500(19)30222-5
Polling
Hargittai Eszter Minh Hao Nguyen Jaelle Fuchs Jonathan Gruber Will Marler Amanda Hunsaker and Goumlkccedile Karaoglu 2020 ldquoCovid-19 Study on Digital Media and the Coronavirus Pandemicrdquo Internet Use and Society Division Institute of Communication and Media Research University of Zurich httpwebuseorgcovid
Navigator Research 2020 ldquoPublic Opinion on Coronavirus Navigator Updaterdquo Navigating Coronavirus (blog) May 21 2020 https navigatorresearchorgnavigating-coronavirus
Russonello Giovanni 2020 ldquoBig Government For Now Most Americans Say Bring It Onrdquo The New York Times May 1 2020 sec US httpswww nytimescom20200501uspoliticscoronavirus-spending-pollshtml
ldquoWashington Post-University of Maryland National Pollrdquo 2020 Washington Post April 21 2020 httpswwwwashingtonpostcomcontext washington-post-university-of-maryland-national-poll-april-21-26-20203583b4e9-66be-4ed6-a457-f6630a550ddf
Kirzinger Ashley Liz Hamel Cailey Muntildeana Audrey Kearney and Mollyann Brodie 2020 ldquoKFF Health Tracking Poll ndash Late April 2020 Coronavirus Social Distancing and Contact Tracingrdquo Kaiser Family Foundation (blog) April 24 2020 httpswwwkfforgcoronavirus-covid-19issue-briefkff-health-tracking-poll-late-april-2020
NYC DOH 2020 ldquoAge-Adjusted Rates of Lab Confirmed COVID-19 Non-hospitalized Cases Estimated Non-Fatal Hospitalized Cases and Persons Known to Have Died per 100000 by RaceEthnicity Grouprdquo New York
118 Resources
City Department of Health httpswww1nycgovassetsdohdownloadspdfimmcovid-19-deaths-race-ethnicity-04242020-1pdf
Anderson Monica and Brooke Auxier 2020 ldquoMost Americans Donrsquot Think Cellphone Tracking Will Help Limit COVID-19 Are Divided on Whether Itrsquos Acceptablerdquo Pew Research Center (blog) April 16 2020 httpswwwpewresearchorgfact-tank20200416most-americans-dont-think-cellphone-tracking-will-help-limit-covid-19-are-divided-on-whether-its-acceptable
Daly Kyle 2020 ldquoExclusive Americans Wary of Giving up Data to Fight Coronavirusrdquo Axios April 3 2020 httpswwwaxioscomexclusive- americans-wary-of-giving-up-data-to-fight-coronavirus-330fc1d9-8b99-4a51-871b-25ee0e0591f2html
Elliott Douglas J Ana Kreacic Lorenzo Milans del Bosch and Lisa Quest nd ldquoData-Sharing in the Time of Coronavirusrdquo Oliver Wyman Forum Accessed May 19 2020 httpswwwoliverwymanforumcomfuture-of-data2020aprdata-sharing-in-the-time-of-coronavirushtml
Auxier Brooke Lee Rainie Monica Anderson Andrew Perrin Madhu Kumar and Erica Turner 2019 ldquoAmericans and Privacy Concerned Confused and Feeling Lack of Control Over Their Personal Informationrdquo Pew Research
Center Internet Science amp Tech (blog) November 15 2019 httpswww pewresearchorginternet20191115americans-and-privacy-concerned- confused-and-feeling-lack-of-control-over-their-personal-information
EPIC 2020 ldquoPublic Opinion on Privacyrdquo Electronic Privacy Information Center January 22 2020 httpsepicorgprivacysurvey
Pew Research Center 2020 ldquoDemographics of Mobile Device Ownership and Adoption in the United Statesrdquo Internet amp Technology Pew Research Cen-ter httpswwwpewresearchorginternetfact-sheetmobile
Rodrigues Rafaela Alina Husain Amanda Couture-Carron Leslye E Orloff and Nawal H Ammar 2018 ldquoPromoting Access to Justice for Immigrant and Limited English Proficient Crime Victims in an Age of Increased Immi-gration Enforcement Initial Report from a 2017 National Surveyrdquo Wash-ington DC National Immigrant Womenrsquos Advocacy Project American University Washington College of Law httpniwaplibrarywclamericaneduwp-contentuploadsImmigrant-Access-to-Justice-National-Reportpdf
Center for Survey Measurement (CSM) MEMORANDUM FOR Associate Directorate for Research and Methodology (ADRM) 2017 ldquoRespondent Confidentiality Concernsrdquo September 20 2017 httpswww2censusgovcacnacmeetings2017-11Memo-Regarding-Respondent-Confidentiality- Concernspdf
Resources 119
Pew Research Center 2017 ldquoUS Muslims Concerned About Their Place in Society but Continue to Believe in the American Dreamrdquo Religion amp Public Life Pew Research Center httpswwwpewforumorg20170726findings-from-pew-research-centers-2017-survey-of-us-muslims
Popular Press
Dixit Pranav 2020 ldquoIndiarsquos Contact Tracing App Is All But Mandatory So This Programmer Hacked It So That He Always Appears Saferdquo BuzzFeed News May 12 2020
Mills Rodrigo Chris 2020 ldquoDigital Contact Tracing Is Becoming Available but Is It Effectiverdquo The Hill May 7 2020 httpsthehillcompolicytechnology 496498-digital-contact-tracing-is-becoming-available-but-is-it-effective
Ingram David 2020 ldquoApple Google Push Makers of Coronavirus Apps Not to Record User Locationrdquo May 4 2020 NBC News Accessed May 19 2020 httpswwwnbcnewscomtechtech-newscoronavirus-apps-won-t-be-able- record-users-location-apple-n1199586
McMinn Sean 2020 ldquoMobile Phone Data Show More Americans Are Leav- ing Their Homes Despite Ordersrdquo NPR May 1 2020 httpswwwnpr org20200501849161820mobile-phone-data-show-more-americans- are-leaving-their-homes-despite-orders
Valentino-DeVries Jennifer Natasha Singer and Aaron Krolik 2020 ldquoA Scram-ble for Virus Apps That Do No Harmrdquo The New York Times April 29 2020 sec Business httpswwwnytimescom20200429 businesscoronavirus-cellphone-apps-contact-tracinghtml
Giglio Mike 2020 ldquoWould You Sacrifice Your Privacy to Get Out of Quaran-tinerdquo The Atlantic April 22 2020 httpswwwtheatlanticcompoliticsarchive202004coronavirus-pandemic-privacy-civil-liberties-911609172
Bradshaw Tim 2020 ldquo2 Billion Phones Cannot Use Google and Apple Con-tact-Tracing Tech | Ars Technicardquo Ars Technica April 20 2020 httpsarstechnicacomtech-policy2020042-billion-phones-cannot-use-google-and-apple-contract-tracing-tech
Reston Maeve Kristina Sgueglia and Cheri Mossburg 2020 ldquoGovernors on East and West Coasts Form Pacts to Decide When to Reopen Economiesrdquo CNN Politics April 13 2020 httpswwwcnncom20200413politicsstates-band-together-reopening-plansindexhtml
Cochrane Emily Claire Cain Miller and Jim Tankersley 2020 ldquoTrump Admin-istration Scales Back Paid Leave in Coronavirus Relief Lawrdquo The New York Times April 2 2020 sec US httpswwwnytimescom20200402uspoliticscoronavirus-paid-leavehtml
120 Resources
Kim Nemo 2020 ldquolsquoMore Scary than Coronavirusrsquo South Korearsquos Health Alerts Expose Private Livesrdquo The Guardian March 6 2020 sec World news httpswwwtheguardiancomworld2020mar06more-scary- than-coronavirus-south-koreas-health-alerts-expose-private-lives
Hickey Matt 2014 ldquoCarriers Can Now Install Apps On Android Handsets Without Customersrsquo Permissionrdquo Forbes December 1 2014 httpswww forbescomsitesmatthickey20141201carriers-can-now-install-apps-on- android-handsets-without-customers-permission
Commentaries
Giubilini Alberto 2020 ldquoContact-Tracing Apps and the Future COVID-19 Vaccination Should Be Compulsory Social Technological and Pharmaco-logical Immunisationrdquo Practical Ethics (blog) May 6 2020 httpblo gpracticalethicsoxacuk202005contact-tracing-apps-and-the-future- covid-19-vaccination-should-be-compulsory-social-technological-and- pharmacological-immunisation
Landau Susan Christy Lopez and Laura Moy 2020 ldquoThe Importance of Equity in Contact Tracingrdquo Lawfare (blog) May 1 2020 httpswww lawfareblogcomimportance-equity-contact-tracing
Schaefer G Owen and Angela Ballantyne 2020 ldquoDownloading COVID-19 Contact Tracing Apps Is a Moral Obligationrdquo Journal of Medical Ethics
Blog (blog) May 4 2020 httpsblogsbmjcommedical-ethics20200504downloading-covid-19-contact-tracing-apps-is-a-moral-obligation
OrsquoNeill Patrick Howell 2020 ldquoFive Things We Need to Do to Make Contact Tracing Really Workrdquo MIT Technology Review April 28 2020 httpswwwtechnologyreviewcom202004281000714five-things-to-make- contact-tracing-work-covid-pandemic-apple-google
Doffman Zak 2020 ldquoCOVID-19 Contact Tracing Why Apple And Google Canrsquot Make This Workrdquo Forbes April 27 2020 httpswwwforbescomsiteszakdoffman20200427this-is-the-contact-tracing-worry-even-apple- and-google-cant-resolve
All Tech is Human 2020 The Ethics of Contact Tracing for COVID-19 httpswwwyoutubecomwatchv=59mKUAVDhdkampt=626s
Canca Cansu 2020 ldquoWhy lsquoMandatory Privacy-Preserving Digital Contact Tracingrsquo Is the Ethical Measure against COVID-19rdquo Medium May 4 2020 httpsmediumcomcansucancawhy-mandatory-privacy-preserv-ing-digital-contact-tracing-is-the-ethical-measure-against-covid-19-a0d-143b7c3b6
Goodman Bryce 2020 ldquoCOVID and Contact Tracing When Social Justice
Resources 121
Demands Mass Surveillancerdquo Medium April 10 2020 httpsmediumcombwgoodmancovid-and-contact-tracing-when-social-justice-demands-mass-surveillance-18d419b8cc5
Gray Rosie and Caroline Haskins 2020 ldquoThey Were Opposed To Govern-ment Surveillance Then The Coronavirus Pandemic Beganrdquo BuzzFeed News March 30 2020 httpswwwbuzzfeednewscomarticlerosiegraythey-were-opposed-to-government-surveillance-then-the
Cegłowski Maciej 2020 ldquoWe Need A Massive Surveillance Programrdquo Idle
Words (blog) March 23 2020 httpsidlewordscom202003we_need_ a_massive_surveillance_programhtm
Academic Literature
COVID-19 Specific
Abeler J Baumlcker M Buermeyer U Zillessen H (2020) COVID-19 Contact Tracing and Data Protection Can Go Together JMIR mHealth and uHealth 8(4) e19359 doi 10219619359
Altmann S Milsom L Zillessen H et al (2020) Acceptability of App-Based Contact Tracing for COVID-19 Cross-Country Survey Evidence Preprint
Berke A Bakker M Vepakomma P Larson K Pentland A (2020) Assessing Disease Exposure Risk with Location Data A Proposal for Cryptographic Preservation of Privacy arXiv arXiv200314412ndashMarch 2020
Bonsail D Parker M Fraser C (2020) Sustainable Containment of COVID-19 Using Smartphones in China Scientific and Ethical Underpinnings for Implementation of Similar Approaches in Other Settings Unpublished working paper
Bradshaw WJ Alley EC Huggins JH Lloyd AL Esvelt KM (2020) Bidirectional Contact Tracing Is Required for Reliable COVID-19 Control Preprint via MedRxiv
Braithwaite I Callender T Bullock M Aldridge R (2020) Automated and Semi-Automated Contact Tracing Protocol for a Rapid Review of Avail-able Evidence and Current Challenges to Inform the Control of COVID-19 Preprint via medRxiv doi httpsdoiorg1011012020041420063636
Bulchandani Bannerjee V Shivam S Moudgalya S Sondhi SL (2020) Digital Herd Immunity and COVID-19 Preprint via medRxiv doi httpsdoiorg1011012020041520066720
Cheng H Jian S Liu D (2020) Contact Tracing Assessment of COVID-19 Transmission Dynamics in Taiwan and Risk at Different Exposure Periods Before and After Symptom Onset JAMA Internal Medicine doi101001jamainternmed20202020
122 Resources
Cho H Ippolito D Yu YW (2020) Contact Tracing Mobile Apps for COVID-19 Privacy Considerations and Related Trade-offs httpsarxivorgpdf200311511pdf
Devakumar D Geordan S Bhopal SS Abubakar I (2020) Racism and dis-crimination in COVID-19 responses The Lancet 395(10231) 1194 doi 101016S0140-6736(20)30792-3
Drew D Nguyen L Steves C et al (2020) Rapid Implementation of Mobile Technology for Real-Time Epidemiology of COVID-19 Science published online May 5 2020
Ferretti L Wymant C Kendall M et al (2020) Quantifying SARS-CoV-2 Trans-mission Suggests Epidemic Control with Digital Contact Tracing Science doi 101126scienceabb6936
Fraser C Abeler-Doumlrner L Ferretti L et al (2020) Digital Contact Tracing Comparing the Capabilities of Centralised and Decentralised Data Archi-tectures to Effectively Suppress the COVID-19 Epidemic While Maximizing Freedom of Movement and Maintaining Privacy Preprint
Leith DJ Farrell S (2020) Coronavirus Contact Tracing Evaluating The Poten-tial Of Using Bluetooth Received Signal Strength For Proximity Detection Preprint
Jayant Limaye R Sauer M Ali J et al (2020) Building Trust While Influencing Online COVID-19 Content iIn the Social Media World The Lancet Digital Health
Mahmood S Hasan K Colder Carras M Labrique A (2020) Global Prepared-ness Aagainst COVID-19 We Must Leverage the Power of Digital Health JMIR Public Health Surveill 20206(2)e18980 DOI 10219618980
Mello M Wang CJ (2020) Ethics and Governance for Digital Disease Surveil-lance Science 11 May 2020 eabb9045 DOI 101126scienceabb9045
Park S Jeehyun Choi G Ko H (2020) Information TechnologyndashBased Tracing Strategy in Response to COVID-19 in South KoreamdashPrivacy Controversies JAMA doi101001jama20206602
Parker M Fraser C Abeler-Doumlrner L Bonsall D (2020) Ethics of Instantaneous Contract Tracing Using Mobile Phone Apps in the Control of the COVID-19 Pandemic Journal of Medical Ethics Published Online May 4 2020
Ethics and Digital Disease Detection
Aiello A Renson A Civich P (2020) Social Mediandash and Internet-Based Dis-ease Surveillance for Public Health Annual Review of Public Health 41 101ndash118 doi 101146annurev-publhealth-040119-094402
Ali J DiStefano M Coates McCall I et al (2019) Ethics of Mobile Phone Sur-
Resources 123
veys to Monitor Non-Communicable Disease Risk Factors in Low- and Middle-Income Countries A Global Stakeholder Survey Global Public Health 14(8) 1167ndash1181
Ali J Labrique A Gionfriddo K et al (2017) Ethics Considerations in Global Mobile Phone-Based Surveys of Noncommunicable Diseases A Concep-tual Exploration Journal of Medical Internet Research 19(5) e110 doi 102196jmir7326
Brownstein J Freifeld C Madoff L (2009) Digital Disease DetectionndashHar-nessing the Web for Public Health Surveillance New England Journal of Medicine 360(21) 2153ndash2157 doi 101056NEJMp0900702
Danquah LO Hasham N MacFarlane M et al (2019) Use of a Mobile Application for Ebola Contact Tracing and Monitoring in Northern Sierra Leone A Proof-of-Concept Study BMC Infectious Diseases 19 810
Degeling C Carter S van Oijen A et al (2020) Community Perspectives on the Benefits and Risks of Technologically Enhanced Communicable Disease Surveillance Systems A Report on Four Community Juries BMC Medical Ethics 21 31 doi 101186s12910-020-00474-6
DeJong B Badou G Luten J et al (2019) Ethical Considerations for Movement Mapping to Identify Disease Transmission Hotspots Emerging Infectious Diseases 25(7) e181421 doi 103201eid2507181421
Denecke K (2017) An Ethical Assessment Model for Digital Disease Detec-tion Technologies Life Sciences Society and Policy 13 16 doi 101186s40504-017-0062-x
Genevieve LD Martani A Wangmo T et al (2019) Participatory Disease Surveillance Systems Ethical Framework Journal of Medical Internet Research 21(5) e12273 doi10219612273
Gilbert G Degeling C Johnson J (2017) Communicable Disease Surveillance Ethics in the Age of Big Data and New Technology Asian Bioethics Review 11 173-187 doi 101007s41649-019-00087-1
Iwaya LH Li J Fischer-Hubner S et al (2019) E-Consent for Data Privacy Consent Management for Mobile Health Technologies in Public Health Surveys and Disease Surveillance Studies in health technology and infor-matics 264 1223-1227 doi 103233SHTI190421
Kostkova P (2018) Disease Surveillance Data Sharing for Public Health The Next Ethical Frontiers Life Sciences Society and Policy 14 16 doi 101186s40504-018-0078-x
Mahmood S Hasan K Colder Carras M Labrique A (2020) Global Prepared-ness against COVID-19 We Must Leverage the Power of Digital Health JMIR Public Health Surveillance 6(2)e18980
124 Resources
Smolinski MS Crawley AW Baltrusaitis K et al (2015) Flu Near You Crowd-sourced Symptom Reporting Spanning 2 Influenza Seasons AJPH 105(10) 2124ndash2130
Wojcik O Brownstein J Chunara R Johansson M (2014) Public Health for the People Participatory Infectious Disease Surveillance in the Digital Age Emerging Themes in Epidemiology 11 7 doi 1011861742-7622-11-7
Other Works
Bernstein J Holroyd TA Atwell JE et al (2019) Rockland Countyrsquos Proposed Ban against Unvaccinated Minors Balancing Disease Control Trust and Liberty Vaccine 37(30) 3933ndash3935
Berry SM Petzold EA Dull P et al (2017) A Response Adaptive Randomiza-tion Platform Trial for Efficient Evaluation of Ebola Virus Treatments A Model for Pandemic Response Clinical Trials 13(1) 22ndash30 doi 101177 1740774515621721
Beukenhorst AL Schultz DM McBeth J (2017) Using Smartphones for Research outside Clinical Settings How Operating Systems App Develop-ers and Users Determine Geolocation Data Quality in mHealth Studies MEDINFO 2017 Precision Healthcare through Informatics
Bourne P (2015) Confronting the Ethical Challenges of Big Data in Public Health Plos Computational Biology 11(2) e1004073 doi 101371journal pcbi1004073
Doerr M Suver C Wilbanks J (2016) Developing a Transparent Par ticipant-Navigated Electronic Informed Consent for Mobile-Mediated Research (April 22 2016) Available at SSRN httpsssrncomabstract=2769129 or httpdxdoiorg102139ssrn2769129
Dredze M Paul MJ Bergsma S Tran H (2013) Carmen A Twitter Geolocation System with Applications to Public Health Expanding the Boundaries of Health Informatics Using Artificial Intelligence Papers from the AAAI 2013 Workshop
Eckhoff PA Tatem AJ (2015) Digital methods in epidemiology can transform disease controlInternational Health Volume 7 Issue 2 March 2015 Pages 77ndash78 httpsdoiorg101093inthealthihv013
Edelstein M Lee L Herten-Crabb A Heymann D Harper D (2018) Strength-ening Global Public Health Surveillance through Data and Benefit Sharing Emerging Infectious Diseases 24(7) 1324ndash1330 doi 103201eid2407151830
Faden R Beauchamp T (1986) A History and Theory of Informed Consent Oxford University Press ISBN 9780199748655
Resources 125
Fairchild A Bayer R (2004) Ethics and the Conduct of Public Health Surveil-lance Science 303(5658) 631ndash632
Flanagan M Howe DC Nissenbaum H (2008) Embodying Values in Technol-ogy Theory and Practice In Information Technology and Moral Philoso-phy van den Hoven J amp Weckert J (eds) Cambridge Cambridge Univer-sity Press 322ndash353
Fraccaro P Beukenhorst A Sperrin M et al (2019) Digital Biomarkers from Geolocation Data in Bipolar Disorder and Schizophrenia A Systematic Review Journal of the American Medical Informatics Association Volume 26 Issue 11 November 2019 Pages 1412ndash1420
Furlanello C Merler S Menegon S et al (2002) New WEBGIS Technologies for Geo-location of Epidemiological Data An Application for the Sur-veillance of the Risk of Lyme borreliosis Disease In Giornale Italiano di Aritmologia e Cardiostimolazione Special Issue Proceedings of the 15th Int Congress ldquoThe New Frontiers of Arrhythmiasrdquo v 5 n1 Mar 2002 241ndash245
Gibson DG Wosu AC Pariyo GW et al (2019) Effect of Airtime Incentives on Response and Cooperation Rates in Non-communicable Disease Interactive Voice Response Surveys Randomised Controlled Trials in Bangladesh and Uganda BMJ Global Health 4(5)
Knobel C Bowker GC (2011) Computing Ethics Values in Design Communi-cations of the acm 54(7) 26ndash28
Lee R Cui RR Muessig KE et al (2015) Incentivizing HIVSTI Testing A Sys-tematic Review of the Literature AIDS and Behavior 18(5) 905ndash912 doi 101007s10461-013-0588-8
Lorchan LT Wyatt J (2014) mHealth and Mobile Medical Apps A Frame-work to Assess Risk and Promote Safer Use Journal of Medical Internet Research 16(9) e210 doi 102196jmir3133
Mathews S McShea M Hanley C et al (2019) Digital Health A Path to Vali-dation Digital Medicine 238 doi 101038s41746-019-0111-3
Moore S Tasse A Thorogood A et al (2017) Consent Processes for Mobile App Mediated Research Systematic Review JMIR mHealth amp uHealth 5(8) e126 DOI 102196mhealth7014
Pallman P Bedding AW Choodari-Oskooei B et al (2018) Adaptive Designs in Clinical Trials Why Use Them and How to Run and Report Them BMC Medicine 16(29)
Rennie S Buchbinder M Juengst E et al (2020) Scraping the Web for Public Health Gains Ethical Considerations from a lsquoBig Datarsquo Research Project on HIV and Incarceration Public Health Ethics doi 101093phephaa006
126 Resources
Rithalia A McDaid C Suekarran S (2009) Impact of Presumed Consent for Organ Donation on Donation Rates A Systematic Review BMJ 2009 338
Singer E Ye C (2012) The Use and Effects of Incentives in Surveys The ANNALS of the American Academy of Political and Social Science 645(1) 112ndash141
Vayena E Blassime A (2018) Health Research with Big Data Time for Systemic Oversight J Law Med Ethics 2018 Mar 46(1) 119ndash129
Vayena E Mastroianni A Kahn J (2012) Ethical Issues in Health Research with Novel Online Sources American Journal of Public Health 102(12) 2225ndash2230 doi 102105AJPH2012300813
Vayena E Salatheacute M Madoff L Brownstein J (2015) Ethical Challenges of Big Data in Public Health Plos Computational Biology 11(2) e1003904 doi 101371journalpcbi1003904
Xafis V Schaefer GO Labude MK et al (2019) An Ethics Framework for Big Data in Health and Research Asian Bioethics Review volume 11 pages 227ndash254(2019)
Works Cited
127
Africa CDC 2020 ldquoGuidance on Contact Tracing for COVID-19 Pandemicrdquo Manuals Guidelines amp Frameworks African Union Africa CDC httpsafricacdcorgdownloadguidance-on-contact-tracing-for-covid-19- pandemic
Aiello Allison E Audrey Renson and Paul N Zivich 2020 ldquoSocial Mediandash and Internet-Based Disease Surveillance for Public Healthrdquo Annual Review of Public Health 41 (1) 101ndash18 httpsdoiorg101146annurev-publhealth-040119-094402
Ali Joseph Michael J DiStefano Iris Coates McCall Dustin G Gibson Gulam Muhammed Al Kibria George W Pariyo Alain B Labrique and Adnan A Hyder 2019 ldquoEthics of Mobile Phone Surveys to Monitor Non-Communicable Disease Risk Factors in Low- and Middle-Income Countries A Global Stakeholder Surveyrdquo Global Public Health 14 (8) 1167ndash81 httpsdoiorg1010801744169220191566482
Ali Joseph Alain B Labrique Kara Gionfriddo George Pariyo Dustin G Gibson Bridget Pratt Molly Deutsch-Feldman and Adnan A Hyder 2017 ldquoEthics Considerations in Global Mobile Phone-Based Surveys of Noncommunicable Diseases A Conceptual Explorationrdquo Journal of Medi-cal Internet Research 19 (5) e110 httpsdoiorg102196jmir7326
Allen Danielle Julius Krein Ganesh Sitaraman and E Glen Weyl 2020 ldquoNational Covid-19 Testing Action Planrdquo The Rockefeller Foundation httpswwwrockefellerfoundationorgwp-contentuploads202004TheRockefellerFoundation_WhitePaper_Covid19_4_22_2020pdf
Altmann Samuel Luke Milsom Hannah Zillessen Raffaele Blasone Frederic Gerdon Ruben Bach Frauke Kreuter Daniele Nosenzo Severine Tous-saert and Johannes Abeler 2020 ldquoAcceptability of App-Based Contact Tracing for COVID-19 Cross-Country Survey Evidencerdquo MedRxiv May 2020050520091587 httpsdoiorg1011012020050520091587
Anderson Monica and Brooke Auxier 2020 ldquoMost Americans Donrsquot Think Cellphone Tracking Will Help Limit COVID-19 Are Divided on Whether
128 Works Cited
Itrsquos Acceptablerdquo Pew Research Center (blog) April 16 2020 https wwwpewresearchorgfact-tank20200416most-americans-dont-think-cellphone-tracking-will-help-limit-covid-19-are-divided-on-whether-its- acceptable
Anderson Monica and Andrew Perrin 2017 ldquoDisabled Americans Less Likely to Use Technologyrdquo Pew Research Center (blog) April 7 2017 httpswwwpewresearchorgfact-tank20170407disabled-americans- are-less-likely-to-use-technology
Apple and Google nd ldquoPrivacy-Preserving Contact Tracingrdquo Apple Accessed May 19 2020 httpswwwapplecomcovid19contacttracing
ASTHO 2012 ldquoPublic Health Collection Use Sharing and Protection of Informationrdquo Issue Brief ASTHO Legal Preparedness Series Association of State and Territorial Health Officials httpswwwasthoorgProgramsPreparednessPublic-Health-Emergency-LawPublic-Health-and- Information-Sharing-ToolkitCollection-Use-Sharing-and-Protection- Issue-Brief
Auxier Brooke Lee Rainie Monica Anderson Andrew Perrin Madhu Kumar and Erica Turner 2019 ldquoAmericans and Privacy Concerned Confused and Feeling Lack of Control Over Their Personal Informationrdquo Pew Research Center Internet Science amp Tech (blog) November 15 2019 httpswwwpewresearchorginternet20191115americans-and-privacy- concerned-confused-and-feeling-lack-of-control-over-their-personal-infor mation
Barth Susanne and Menno D T de Jong 2017 ldquoThe Privacy Paradox ndash Inves-tigating Discrepancies between Expressed Privacy Concerns and Actual Online Behavior ndash A Systematic Literature Reviewrdquo Telematics and Infor-matics 34 (7) 1038ndash58 httpsdoiorg101016jtele201704013
Bernstein Justin Taylor A Holroyd Jessica E Atwell Joseph Ali and Rupali J Limaye 2019 ldquoRockland Countyrsquos Proposed Ban against Unvaccinated Minors Balancing Disease Control Trust and Libertyrdquo Vaccine June httpsdoiorg101016jvaccine201905093
Berry Scott M Elizabeth A Petzold Peter Dull Nathan M Thielman Coleen K Cunningham G Ralph Corey Micah T McClain et al 2016 ldquoA Response Adaptive Randomization Platform Trial for Efficient Evalua-tion of Ebola Virus Treatments A Model for Pandemic Responserdquo Clinical Trials (London England) 13 (1) 22ndash30 httpsdoiorg101177 1740774515621721
Beukenhorst AL DM Schultz J McBeth R Lakshminarayana JC Sergeant and WG Dixon 2017 ldquoUsing Smartphones for Research Outside Clinical
Works Cited 129
Settings How Operating Systems App Evelopers and Users Determine Geolocation Data Quality in MHealth Studiesrdquo In MEDINFO 2017 Precision Healthcare through Infomatics 10ndash14 IOS Press httpebooks iospressnlvolumearticle48095
Bradshaw Tim 2020 ldquo2 Billion Phones Cannot Use Google and Apple Con-tact-Tracing Tech | Ars Technicardquo Ars Technica April 20 2020 httpsarstechnicacomtech-policy2020042-billion-phones-cannot-use-google-and-apple-contract-tracing-tech
Canca Cansu 2020 ldquoWhy lsquoMandatory Privacy-Preserving Digital Contact Tracingrsquo Is the Ethical Measure against COVID-19rdquo Medium May 4 2020 httpsmediumcomcansucancawhy-mandatory-privacy- preserving-digital-contact-tracing-is-the-ethical-measure-against- covid-19-a0d143b7c3b6
Cavalier Robert J ed 2011 Approaching Deliberative Democracy Theory and Practice Carnegie Mellon University Press
Cavoukian Ann 2010 ldquoPrivacy by Design The 7 Foundational Principlesrdquo Ontario Canada Information and Privacy Commissioner of Canada httpsiapporgmediapdfresource_centerpbd_implement_7found_ principlespdf
CDC 2005 ldquoVI Plan for Surveillance of Contacts of SARS Case Supplement B SARS Surveillancerdquo Centers for Disease Control and Prevention httpswwwcdcgovsarsguidanceb-surveillancecontactshtml
mdashmdashmdash 2020a ldquoContact Tracingrdquo Get and Keep America Open Supporting States Tribes Localities and Territories Centers for Disease Control and Prevention httpswwwcdcgovcoronavirus2019-ncovphpopen-americacontact-tracinghtml
mdashmdashmdash 2020b ldquoContact Tracing Part of a Multipronged Approach to Fight the COVID-19 Pandemicrdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention httpswwwcdcgovcoronavirus2019- ncovphpprinciples-contact-tracinghtml
mdashmdashmdash 2020c ldquoInterim Guidance for Businesses and Employersrdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention httpswwwcdcgovcoronavirus2019-ncovcommunityguidance-business- responsehtml
mdashmdashmdash 2020d ldquoOperational Considerations for the Identification of Healthcare Workers and Inpatients with Suspected COVID-19 in Non-US Healthcare Settingsrdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention httpswwwcdcgovcoronavirus2019-ncovhcpnon-us-settingsguidance-identify-hcw-patientshtml
130 Works Cited
mdashmdashmdash 2020e ldquoPreliminary Criteria for the Evaluation of Digital Contact Trac-ing Tools for COVID-19rdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention httpswwwcdcgovcoronavirus 2019-ncovdownloadsphpprelim-eval-criteria-digital-contact-tracingpdf
mdashmdashmdash 2020f ldquoLegal Authorities for Isolation and Quarantinerdquo Centers for Disease Control and Prevention httpswwwcdcgovquarantine aboutlawsregulationsquarantineisolationhtml
mdashmdashmdash 2020g ldquoDiscontinuation of Isolation for Persons with COVID-19 Not in Healthcare Settingsrdquo Interim Guidance Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention httpswww cdcgovcoronavirus2019-ncovhcpdisposition-in-home-patientshtml
mdashmdashmdash 2020h ldquoCOVID-19 Provisional CountsndashWeekly Updates by Select Demographic and Geographic Characteristicsrdquo CDC National Vital Statis-tics System httpswwwcdcgovnchsnvssvsrrcovid_weeklyindexhtm
Center for Survey Measurement (CSM) MEMORANDUM FOR Associate Directorate for Research and Methodology (ADRM) 2017 ldquoRespondent Confidentiality Concernsrdquo September 20 2017 httpswww2censusgovcacnacmeetings2017-11Memo-Regarding-Respondent-Confidentiality- Concernspdf
Cheng Hao-Yuan Shu-Wan Jian Ding-Ping Liu Ta-Chou Ng Wan-Ting Huang and Hsien-Ho Lin 2020 ldquoContact Tracing Assessment of COVID-19 Transmission Dynamics in Taiwan and Risk at Different Exposure Periods Before and After Symptom Onsetrdquo JAMA Internal Medicine May httpsdoiorg101001jamainternmed20202020
Cochrane Emily Claire Cain Miller and Jim Tankersley 2020 ldquoTrump Admin-istration Scales Back Paid Leave in Coronavirus Relief Lawrdquo The New York Times April 2 2020 sec US httpswwwnytimescom20200402uspoliticscoronavirus-paid-leavehtml
Crocker Andrew Kurt Opsahl and Bennett Cyphers 2020 ldquoThe Challenge of Proximity Apps For COVID-19 Contact Tracingrdquo Electronic Fron-tier Foundation April 10 2020 httpswwwefforgdeeplinks202004challenge-proximity-apps-covid-19-contact-tracing
Danquah Lisa O Nadia Hasham Matthew MacFarlane Fatu E Conteh Fatoma Momoh Andrew A Tedesco Amara Jambai David A Ross and Helen A Weiss 2019 ldquoUse of a Mobile Application for Ebola Contact Tracing and Monitoring in Northern Sierra Leone A Proof-of-Concept Studyrdquo BMC Infectious Diseases 19 (1) 810 httpsdoiorg101186s12879-019-4354-z
de Jong Bouke C Badou M Gaye Jeroen Luyten Bart van Buitenen Emman-
Works Cited 131
uel Andreacute Conor J Meehan Cian OrsquoSiochain et al 2019 ldquoEthical Considerations for Movement Mapping to Identify Disease Transmission Hotspotsrdquo Emerging Infectious Diseases 25 (7) httpsdoiorg103201eid2507181421
Devakumar Delan Geordan Shannon Sunil S Bhopal and Ibrahim Abu-bakar 2020 ldquoRacism and Discrimination in COVID-19 Responsesrdquo Lancet (London England) 395 (10231) 1194 httpsdoiorg101016S0140-6736(20)30792-3
Dixit Pranav 2020 ldquoIndiarsquos Contact Tracing App Is All But Mandatory So This Programmer Hacked It So That He Always Appears Saferdquo BuzzFeed News May 12 2020 httpswwwbuzzfeednewscomarticlepranavdixitindia-aarogya-setu-hacked
Doerr Megan Christine Suver and John Wilbanks 2016 ldquoDeveloping a Trans-parent Participant-Navigated Electronic Informed Consent for Mobile- Mediated Researchrdquo SSRN Scholarly Paper ID 2769129 Rochester NY Social Science Research Network httpsdoiorg102139ssrn2769129
Dredze Mark Michael J Paul Shane Bergsma and Hieu Tran 2013 ldquoCarmen A Twitter Geolocation System with Applications to Public Healthrdquo AAAI Workshops Workshops at the Twenty-Seventh AAAI Conference on Artifi-cial Intelligence httpswwwaaaiorgocsindexphpWSAAAIW13paperview70856497
Eckhoff Philip A and Andrew J Tatem 2015 ldquoDigital Methods in Epidemi-ology Can Transform Disease Controlrdquo International Health 7 (2) 77ndash78 httpsdoiorg101093inthealthihv013
EEOC 2020 ldquoWhat You Should Know About COVID-19 and the ADA the Rehabilitation Act and Other EEO Lawsrdquo US Equal Employment Opportunity Commission May 7 2020 httpswwweeocgovwyskwhat-you-should-know-about-covid-19-and-ada-rehabilitation-act-and-other-eeo-laws
Electronic Privacy Information Center Testimony to Congress 2020 ldquoEPIC to Congress Establish Privacy Safeguards for Digital Contact Tracingrdquo April 15 2020 httpsepicorgtestimonycongressEPIC-HEC-Contact-Tracing-Apr2020pdf
Elliott Douglas J Ana Kreacic Lorenzo Milans del Bosch and Lisa Quest 2020 ldquoData-Sharing in the Time of Coronavirusrdquo Oliver Wyman Forum Accessed May 19 2020 httpswwwoliverwymanforumcomfuture-of-data2020aprdata-sharing-in-the-time-of-coronavirushtml
Faden Ruth R and Tom L Beauchamp 1986 A History and Theory of Informed Consent Oxford University Press
132 Works Cited
FCC and FTC 2017 ldquoFCC-FTC Consumer Protection Memorandum of Understandingrdquo Federal Communications Commission and Federal Trade Commission httpswwwftcgovsystemfilesdocumentscooperation_agreements151116ftcfcc-moupdf
Ferretti Luca Chris Wymant Michelle Kendall Lele Zhao Anel Nurtay Lucie Abeler-Doumlrner Michael Parker David Bonsall and Christophe Fraser 2020 ldquoQuantifying SARS-CoV-2 Transmission Suggests Epidemic Control with Digital Contact Tracingrdquo Science 368 (6491) httpsdoiorg101126scienceabb6936
Fishkin James S and Peter Laslett eds 2003 Debating Deliberative Democ-racy Blackwell Publisher Ltd httpsonlinelibrarywileycomdoibook 1010029780470690734
Flanagan Mary Daniel C Howe and Helen Nissenbaum 2008 ldquoEmbodying Values in Technology Theory and Practicerdquo In Information Technology and Moral Philosophy edited by Jeroen van den Hoven and John Weckert 322ndash53 Cambridge University Press
Fraccaro Paolo Anna Beukenhorst Matthew Sperrin Simon Harper Jasper Palmier-Claus Shocircn Lewis Sabine N Van der Veer and Niels Peek 2019 ldquoDigital Biomarkers from Geolocation Data in Bipolar Disorder and Schizophrenia A Systematic Reviewrdquo Journal of the American Medical Informatics Association 26 (11) 1412ndash20 httpsdoiorg101093jamiaocz043
Fraser Christophe Lucie Abeler-Doumlrner Luca Ferretti Michael Parker Michelle Kendall and David Bonsall 2020 ldquoDigital Contact Tracing Comparing the Capabilities of Centralised and Decentralised Data Architectures to Effectively Suppress the COVID-19 Epidemic Whilst Maximising Freedom of Movement and Maintaining Privacyrdquo httpsgithubcomBDI-pathogenscovid-19_instant_tracingblobmasterCentralised20and20decen-tralised20systems20for20contact20tracingpdf
FTC 2012 ldquoProtecting Consumer Privacy in an Era of Rapid Changerdquo FTC Report Recommendations for Businesses and Policymakers Federal Trade Commission httpswwwftcgovsitesdefaultfilesdocumentsreports federal-trade-commission-report-protecting-consumer-privacy- era-rapid-change-recommendations120326privacyreportpdf
mdashmdashmdash 2020 ldquoPrivacy amp Data Security Update 2019rdquo Federal Trade Commis-sion httpswwwftcgovsystemfilesdocumentsreportsprivacy-data- security-update-20192019-privacy-data-security-report-508pdf
Furlanello Cesare Stefano Merler Stefano Menegon Sebastiano Mancuso and Gianni Bertiato 2002 ldquoNew WEBGIS Technologies for Geo-Location of
Works Cited 133
Epidemiological Data An Application for the Surveillance of the Risk of Lyme Borreliosis Diseaserdquo Giornale Italiano Di Aritmologia e Cardiosti-molazione Special Issue Proceedings of the 15th Int Congress ldquoThe New Frontiers of Arrhythmiasrdquo 5 (1) 241ndash45
Gan Nectar and David Culver 2020 ldquoChina Is Fighting the Coronavirus with a Digital QR Code Herersquos How It Worksrdquo CNN Business April 16 2020 httpswwwcnncom20200415asiachina-coronavirus-qr-code-intl-hnkindexhtml
Gibson Dustin G Adaeze C Wosu George William Pariyo Saifuddin Ahmed Joseph Ali Alain B Labrique Iqbal Ansary Khan Elizeus Rutebemberwa Meerjady Sabrina Flora and Adnan A Hyder 2019 ldquoEffect of Airtime Incentives on Response and Cooperation Rates in Non-Communicable Disease Interactive Voice Response Surveys Randomised Controlled Trials in Bangladesh and Ugandardquo BMJ Global Health 4 (5) e001604 httpsdoiorg101136bmjgh-2019-001604
Giglio Mike 2020 ldquoWould You Sacrifice Your Privacy to Get Out of Quarantinerdquo The Atlantic April 22 2020 httpswwwtheatlanticcompoliticsarchive202004coronavirus-pandemic-privacy-civil-liberties-911 609172
Guariglia Matthew 2020 ldquoThe Dangers of COVID-19 Surveillance Proposals to the Future of Protestrdquo Electronic Frontier Foundation April 29 2020 httpswwwefforgdeeplinks202004some-covid-19-surveillance-propos als-could-harm-free-speech-after-covid-19
Hadavas Chloe 2020 ldquoHow Effective Are Contact Tracing Appsrdquo Slate Maga-zine May 13 2020 httpsslatecomtechnology202005contact- tracing-apps-less-effective-icelandhtml
Hamilton Isobel Asher 2020 ldquoPoland Made an App That Forces Coronavirus Patients to Take Regular Selfies to Prove Theyrsquore Indoors or Face a Police Visitrdquo Business Insider March 23 2020 httpswwwbusinessinsidercompoland-app-coronavirus-patients-mandaotory-selfie-2020-3
Hargittai Eszter Minh Hao Nguyen Jaelle Fuchs Jonathan Gruber Will Marler Amanda Hunsaker and Goumlkccedile Karaoglu 2020 ldquoCovid-19 Study on Digital Media and the Coronavirus Pandemicrdquo Internet Use and Society Division Institute of Communication and Media Research University of Zurich httpwebuseorgcovid
Hargittai Eszter and Elissa Redmiles 2020 ldquoWill Americans Be Willing to Install COVID-19 Tracking Appsrdquo Scientific American Blog Net-work April 28 2020 httpsblogsscientificamericancomobservationswill-americans-be-willing-to-install-covid-19-tracking-apps
134 Works Cited
Hart Vi Divya Siddarth Bethan Cantrell Lila Tretikov Peter Eckersley John Langford Scott Leibrand et al 2020 ldquoOutpacing the Virus Dig-ital Response to Containing the Spread of COVID-19 While Mitigat-ing Privacy Risksrdquo Whitepaper 5 COVID-19 Rapid Response Impact Initiative Edmond J Safra Center for Ethics httpsdrivegooglecomfiled1vIN2AX-DDNW-S0aHq8xs0RJ2jkR_CckXview
Hemming K T P Haines P J Chilton A J Girling and R J Lilford 2015 ldquoThe Stepped Wedge Cluster Randomised Trial Rationale Design Anal-ysis and Reportingrdquo BMJ 350 (February) httpsdoiorg101136bmjh391
Hendrix Steve and Ruth Eglash 2020 ldquoIsrael Is Using Cellphone Surveil-lance to Warn Citizens You May Already Be Infectedrdquo Washington Post Accessed May 19 2020 httpswwwwashingtonpostcomworldmiddle_eastisrael-is-using-cellphone-surveillance-to-warn-citizens-you-may- already-be-infected2020031968267294-69e7-11ea-b199-3a9799c54 512_storyhtml
Heneghan Carl Jon Brassey and Tom Jefferson 2020 ldquoCOVID-19 What Proportion Are Asymptomaticrdquo Centre for Evidence-Based Medi-cine httpswwwcebmnetcovid-19covid-19-what-proportion-are- asymptomatic
HHS 2019 ldquoPublic Health Emergency Declarationrdquo Public Health Emergency Accessed May 19 2020 httpswwwphegovPreparednesslegalPagesphedeclarationaspx
Hickey Matt 2014 ldquoCarriers Can Now Install Apps On Android Handsets Without Customersrsquo Permissionrdquo Forbes December 1 2014 httpswww forbescomsitesmatthickey20141201carriers-can-now-install- apps-on-android-handsets-without-customers-permission
Hinch Robert Will Probert Anel Nurtay Michelle Kendall Chris Wymant Matthew Hall Katrina Lythgoe et al 2020 ldquoEffective Configurations of a Digital Contact Tracing App A Report to NHSXrdquo fileUsersameliahoodDownloadsReport20-20Effective20App20Configurations20(1)pdf
Ingram David 2020 ldquoApple Google Push Makers of Coronavirus Apps Not to Record User Locationrdquo May 4 2020 NBC News Accessed May 19 2020 httpswwwnbcnewscomtechtech-newscoronavirus-apps-won-t-be-able- record-users-location-apple-n1199586
Iwaya Leonardo H Jane Li Simone Fischer-Huumlbner Rose-Mharie Aringhlfeldt and Leonardo A Martucci 2019 ldquoE-Consent for Data Privacy Consent Management for Mobile Health Technologies in Public Health Surveys and
Works Cited 135
Disease Surveillancerdquo Studies in Health Technology and Informatics 264 (August) 1223ndash27 httpsdoiorg103233SHTI190421
ldquoJoint Statement on Contact Tracingrdquo 2020 April 19 httpscryptobriefingcomwp-contentuploads202004Joint-Statement-from-Researcherspdf
Kahn Gilmor Daniel 2020 ldquoPrinciples for Technology-Assisted Contact- Tracingrdquo White Paper American Civil Liberties Union httpswwwaclu orgreportaclu-white-paper-principles-technology-assisted-contact-tracing
Kim Max S 2020 ldquoSeoulrsquos Radical Experiment in Digital Contact Tracingrdquo The New Yorker April 17 2020 httpswwwnewyorkercomnewsnews-deskseouls-radical-experiment-in-digital-contact-tracing
Kim Nemo 2020 ldquolsquoMore Scary than Coronavirusrsquo South Korearsquos Health Alerts Expose Private Livesrdquo The Guardian March 6 2020 sec World news httpswwwtheguardiancomworld2020mar06more-scary- than-coronavirus-south-koreas-health-alerts-expose-private-lives
Kirzinger Ashley Liz Hamel Cailey Muntildeana Audrey Kearney and Mollyann Brodie 2020 ldquoKFF Health Tracking Poll ndash Late April 2020 Coronavirus Social Distancing and Contact Tracingrdquo Kaiser Family Foundation (blog) April 24 2020 httpswwwkfforgcoronavirus-covid-19issue-briefkff-health-tracking-poll-late-april-2020
Knobel Cory and Geoffrey C Bowker 2011 ldquoComputing Ethics Value in Designrdquo Communications of the ACM 54 (7) 26ndash28 httpsdoiorgdoi10114519657241965735
Lee Ramon Rosa R Cui Kathryn E Muessig Harsha Thirumurthy and Joseph D Tucker 2014 ldquoIncentivizing HIVSTI Testing A Systematic Review of the Literaturerdquo AIDS and Behavior 18 (5) 905ndash12 https doiorg101007s10461-013-0588-8
Leprince-Ringuet Daphne 2020 ldquoContact-Tracing Apps Why the NHS Said No to Apple and Googlersquos Planrdquo ZDNet April 28 2020 httpswwwzdnetcomarticlecontact-tracing-apps-why-the-nhs-said-no-to-apple-and- googles-plan
Lovejoy Ben 2020 ldquoNHS Contact Tracing App Hands-on as the Test Goes Liverdquo 9to5Mac (blog) May 7 2020 https9to5maccom20200507nhs-contact-tracing
Mahmood Sultan Khaled Hasan Michelle Colder Carras and Alain Labri-que 2020 ldquoGlobal Preparedness Against COVID-19 We Must Leverage the Power of Digital Healthrdquo JMIR Public Health and Surveillance 6 (2) e18980 httpsdoiorg10219618980
Mathews Simon C Michael J McShea Casey L Hanley Alan Ravitz Alain B Labrique and Adam B Cohen 2019 ldquoDigital Health A Path to
136 Works Cited
Validationrdquo Npj Digital Medicine 2 (1) 1ndash9 httpsdoiorg101038s41746-019-0111-3
Mello By Michelle M and C Jason Wang 2020 ldquoEthics and Governance for Digital Disease Surveillancerdquo Science May httpsdoiorg101126science abb9045
Mills Rodrigo Chris 2020 ldquoDigital Contact Tracing Is Becoming Available but Is It Effectiverdquo The Hill May 7 2020 httpsthehillcompolicytechnolo-gy496498-digital-contact-tracing-is-becoming-available-but-is-it-effective
MIT nd ldquoPrivate Kit Safe Paths Privacy-by-Design Covid19 Solutions Using GPS+Bluetooth for Citizens and Public Health Officialsrdquo Safepaths Accessed May 20 2020 httpssafepathsmitedu
ldquoMobile Cellular Subscriptions (per 100 People)rdquo 2018 World Telecommuni-cationICT Development Report International Telecommunication Union httpsdataworldbankorgindicatorITCELSETSP2
Moore Sarah Anne-Marie Tasseacute Adrian Thorogood Ingrid Winship Marsquon Zawati and Megan Doerr 2017 ldquoConsent Processes for Mobile App Mediated Research Systematic Reviewrdquo JMIR MHealth and UHealth 5 (8) e126 httpsdoiorg102196mhealth7014
Morse Jack 2020 ldquoNorth Dakota Launched a Contact-Tracing App Itrsquos Not Going Wellrdquo Mashable Accessed May 19 2020 httpsmashablecomarticlenorth-dakota-contact-tracing-app
Muller Robert T 2020 ldquoCOVID-19 Brings a Pandemic of Conspiracy Theo-riesrdquo Psychology Today April 24 2020 httpswwwpsychologytodaycomblogtalking-about-trauma202004covid-19-brings-pandemic-conspiracy- theories
Mulligan Stephen P and Chris D Linebaugh 2019 ldquoData Protection Law An Overviewrdquo R45631 Congressional Research Service httpscrsreports congressgovproductpdfRR45631
National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research 1979 ldquoThe Belmont Report Ethical Principles and Guidelines for the Protection of Human Subjects of Researchrdquo US Dept of Health amp Human Services Office of Human Research Protections httpswwwhhsgovohrpregulations-and-policybelmont-reportread-the-belmont-reportindexhtml
NDDoH 2020 ldquoNorth Dakota Launches Care19 App to Combat COVID-19rdquo North Dakota Department of Health April 7 2020 httpswwwhealth ndgovnewsnorth-dakota-launches-care19-app-combat-covid-19
NYC DOH 2020 ldquoAge-Adjusted Rates of Lab Confirmed COVID-19 Non-hospitalized Cases Estimated Non-Fatal Hospitalized Cases and Persons Known to Have Died per 100000 by RaceEthnicity Grouprdquo New York
Works Cited 137
City Department of Health httpswww1nycgovassetsdohdownloadspdfimmcovid-19-deaths-race-ethnicity-04242020-1pdf
OrsquoRielly Michael 2018 ldquoFCC Regulatory Free Arenardquo Federal Communica-tions Commission June 1 2018 httpswwwfccgovnews-eventsblog 20180601fcc-regulatory-free-arena
OSHA 2020 ldquoGuidance on Preparing Workplaces for COVID-19rdquo OSHA 3990-03 2020 Department of Labor Occupational Safety and Health Act of 1970 httpswwwoshagovPublicationsOSHA3990pdf
Pallmann Philip Alun W Bedding Babak Choodari-Oskooei Munyaradzi Dimairo Laura Flight Lisa V Hampson Jane Holmes et al 2018 ldquoAdaptive Designs in Clinical Trials Why Use Them and How to Run and Report Themrdquo BMC Medicine 16 (1) 29 httpsdoiorg101186s12916-018-1017-7
Pew Research Center 2017 ldquoUS Muslims Concerned About Their Place in Society but Continue to Believe in the American Dreamrdquo Religion amp Public Life Pew Research Center httpswwwpewforumorg20170726findings-from-pew-research-centers-2017-survey-of-us-muslims
mdashmdashmdash 2020 ldquoDemographics of Mobile Device Ownership and Adoption in the United Statesrdquo Internet amp Technology Pew Research Center Accessed May 19 2020 httpswwwpewresearchorginternetfact-sheetmobile
PHLS 2002 ldquoPrinciples of the Ethical Practice of Public Healthrdquo Public Health Leadership Society httpswwwaphaorg-mediafilespdfmembergroupsethicsethics_brochureashx
PIH 2020a ldquoPart I Testing Contact Tracing and Community Management of COVID-19rdquo PIH Guide | COVID-19 STOP COVID Partners in Health httpswwwpihorgsitesdefaultfiles2020-04PIH_Guide_COVID_Part_I_Testing_Tracing_Community_Managment_4_4pdf
mdashmdashmdash 2020b ldquoBox It Inrdquo PIH Guide | COVID-19 Partners in Health httpspreventepidemicsorgcovid19resourcesbox-it-in
Resolve to Save Lives 2020 ldquoCOVID-19 Contact Tracing Playbookrdquo Vital Strategies httpscontacttracingplaybookresolvetosavelivesorg
Reston Maeve Kristina Sgueglia and Cheri Mossburg 2020 ldquoGovernors on East and West Coasts Form Pacts to Decide When to Reopen Economiesrdquo CNN Politics April 13 2020 httpswwwcnncom20200413politicsstates-band-together-reopening-plansindexhtml
Rithalia Amber Catriona McDaid Sara Suekarran Lindsey Myers and Amanda Sowden 2009 ldquoImpact of Presumed Consent for Organ Donation on Donation Rates A Systematic Reviewrdquo BMJ 338 (January) httpsdoiorg101136bmja3162
Rodrigues Rafaela Alina Husain Amanda Couture-Carron Leslye E Orloff
138 Works Cited
and Nawal H Ammar 2018 ldquoPromoting Access to Justice for Immigrant and Limited English Proficient Crime Victims in an Age of Increased Immi-gration Enforcement Initial Report from a 2017 National Surveyrdquo Wash-ington DC National Immigrant Womenrsquos Advocacy Project American University Washington College of Law httpniwaplibrarywclamericaneduwp-contentuploadsImmigrant-Access-to-Justice-National-Reportpdf
Simpson Erin and Adam Conner 2020 ldquoDigital Contact Tracing to Contain the Coronavirusrdquo Center for American Progress httpswww americanprogressorgissuestechnology-policynews20200422483521digital-contact-tracing-contain-coronavirus
Singer Eleanor and Cong Ye 2013 ldquoThe Use and Effects of Incentives in Surveysrdquo The ANNALS of the American Academy of Political and Social Science 645 (1) 112ndash41 httpsdoiorg1011770002716212458082
Thornton Rebecca L 2008 ldquoThe Demand for and Impact of Learning HIV Statusrdquo The American Economic Review 98 (5) 1829ndash63 httpsdoi org101257aer9851829
US DOE 2009 ldquoState Regulation of Private Schoolsrdquo US Dept of Education Office of Non-Public Education httpswww2edgovadminscommchoiceregprivschlregprivschlpdf
Valentino-DeVries Jennifer 2020 ldquoCellphone Carriers Face $200 Million Fine for Not Protecting Location Datardquo The New York Times February 28 2020 sec Technology httpswwwnytimescom20200228technologyfcc-cellphones-location-data-fineshtml
Valentino-DeVries Jennifer Natasha Singer and Aaron Krolik 2020 ldquoA Scramble for Virus Apps That Do No Harmrdquo The New York Times April 29 2020 sec Business httpswwwnytimescom20200429businesscoronavirus-cellphone-apps-contact-tracinghtml
Washington PostndashUMD ldquoWashington Post-University of Maryland National Pollrdquo 2020 Washington Post April 21 2020 httpswwwwashingtonpostcomcontextwashington-post-university-of-maryland-national-poll-april-21-26-20203583b4e9-66be-4ed6-a457-f6630a550ddf
Watson Crystal Anita Cicero James Blumenstock and Michael Fraser 2020 ldquoA National Plan to Enable Comprehensive COVID-19 Case Finding and Contact Tracing in the USrdquo Johns Hopkins Bloomberg School of Public Health Center for Health Security and the Association of State and Terri-torial Health Officials httpswwwcenterforhealthsecurityorgour-workpubs_archivepubs-pdfs2020200410-national-plan-to-contact-tracingpdf
Works Cited 139
WHO 2017 ldquoWHO Guidelines on Ethical Issues in Public Health Surveillancerdquo World Health Organization httpswwwwhointethicspublicationspublic-health-surveillanceen
mdashmdashmdash 2020 ldquoCoronavirus Disease 2019 (COVID-19)rdquo Situation Report 73 World Health Organization httpswwwwhointdocsdefault-sourcecoro-navirusesituation-reports20200402-sitrep-73-covid-19pdf
- _iy90yrtqgtxz
- _GoBack
- Acknowledgments
- Preface
- Lead Authors and Contributors
- Acronyms and Abbreviations
- Summary
-
- Introduction
- DCTT Features Functions and Potential Applications
- Summary of Recommendations
- Summary of Analysis
-
- Introduction
-
- Guiding Principles for the Use of Digital Public Health Technologies for Pandemic Response
-
- Public Health Perspective
-
- Types of Information Collected through Contact Tracing
- How Contact Tracing Information Informs Public Health Action
- Characteristics That Make Data Useful to Public Health for Reducing Disease Transmission
-
- Digital Technology and Contact Tracing
-
- Characteristics of SARS-CoV-2 Relevant to Candidate Digital Solutions
- Previously Existing Contact Tracing Technologies
- Introduction of Novel Digital Contact Tracing Technologies
- Relevant Differences between Manual and Digital Contact Tracing
-
- Ethics of Designing and Using DCTT
-
- Justifying the Use of DCTT Systems
- Monitoring and Evaluating Technologies to Inform Policy and Practice
- Public Trust and Public Attitudes
- Designing Flexible Technology to Maximize Public Health Utility While Respecting Other Values
- Policy Positions to Advance Widespread Use of Digital Contact Tracing Technologies
- Disclosure and AuthorizationConsent
- Promoting Equity and Fairness
- Instituting Transparent Governance and Oversight
-
- Legal Considerations
-
- Data Privacy and Data Security Laws
- Health Information Privacy
- Labor and Employment Privacy Rights
- Constitutional Privacy Rights
- Consent
- Anti-discrimination and Individual Freedom Laws
-
- Recommendations
-
- Public Health
- Ethics
- Legislative
-
- Resources
-
- US Government Response
- Other Governmental and Nongovernmental Organizations
- Digital Contact Tracing Experiences from Other Countries
- Specific Digital ProductsApps
- Polling
- Popular Press
- Commentaries
- Academic Literature
-
- Works Cited
-
vi Contents
3 Ethics of Designing and Using DCTT 43
Justifying the Use of DCTT Systems 44
Monitoring and Evaluating Technologies to Inform Policy and Practice 45
Public Trust and Public Attitudes 48
Designing Flexible Technology to Maximize Public Health Utility
While Respecting Other Values 50
Policy Positions to Advance Widespread Use of Digital Contact
Tracing Technologies 59
Disclosure and AuthorizationConsent 63
Promoting Equity and Fairness 69
Instituting Transparent Governance and Oversight 72
4 Legal Considerations 75
Data Privacy and Data Security Laws 77
Health Information Privacy 82
Labor and Employment Privacy Rights 84
Constitutional Privacy Rights 86
Consent 93
Anti-discrimination and Individual Freedom Laws 94
5 Recommendations 97
Public Health 97
Ethics 98
Legislative 102
Resources 103
US Government Response 103
Other Governmental and Nongovernmental Organizations 106
Digital Contact Tracing Experiences from Other Countries 112
Specific Digital ProductsApps 115
Polling 117
Popular Press 119
Commentaries 120
Academic Literature 121
Works Cited 127
Lead Authors and Contributors
vii
Lead Authors
Joseph Ali JD Assistant Professor Department of International Health Johns Hopkins Bloomberg School of Public Health Core Faculty amp Associate Director for Global Programs Johns Hopkins Berman Institute of Bioethics
Anne Barnhill PhD Core Faculty amp Research Scholar Johns Hopkins Berman Institute of Bioethics
Anita Cicero JD Deputy Director Johns Hopkins Center for Health Security Visiting Faculty Johns Hopkins Bloomberg School of Public Health
Katelyn Esmonde PhD Hecht-Levi Postdoctoral Fellow Johns Hopkins Berman Institute of Bioethics
Amelia Hood MA Research Program Coordinator Johns Hopkins Berman Insti-tute of Bioethics
Brian Hutler PhD JD Hecht-Levi Postdoctoral Fellow Johns Hopkins Berman Institute of Bioethics
Jeffrey Kahn PhD MPH Andreas C Dracopoulos Director Johns Hopkins Ber-man Institute of Bioethics
Alan Regenberg MBE Director of Outreach amp Research Support Associate Fac-ulty Johns Hopkins Berman Institute of Bioethics
Crystal Watson DrPH MPH Senior Scholar Johns Hopkins Center for Health Security Assistant Professor Department of Environmental Health and Engi-neering Johns Hopkins Bloomberg School of Public Health
Matthew Watson Senior Analyst Johns Hopkins Center for Health Security Senior Research Associate Department of Environmental Health and Engi-neering Johns Hopkins Bloomberg School of Public Health
Other Contributors
Robert Califf MD MACC Head of Clinical Policy and Strategy Verily and Goo-gle Health
Ruth Faden PhD MPH Philip Franklin Wagley Professor of Biomedical Ethics amp Founder Johns Hopkins Berman Institute of Bioethics
viii Lead Authors and Contributors
Divya Hosangadi MSPH Senior Analyst Johns Hopkins Center for Health Secu-rity Research Associate Department of Environmental Health and Engi-neering Johns Hopkins Bloomberg School of Public Health
Nancy Kass ScD Deputy Director for Public Health amp Phoebe R Berman Pro-fessor of Bioethics and Public Health Johns Hopkins Berman Institute of Bioethics
Alain Labrique PhD MHS MS Professor Department of International Health Johns Hopkins Bloomberg School of Public Health Director Johns Hopkins Global Health Initiative
Deven McGraw JD MPH LLM Chief Regulatory Officer CiitizenMichelle Mello JD PhD Professor of Law Stanford Law School Professor of
Health Research and Policy Stanford University School of MedicineMichael Parker BEd (Hons) MA PhD Director Wellcome Centre for Ethics and
Humanities Ethox Centre University of OxfordStephen Ruckman JD MSc MAR Senior Advisor to the President for Policy
Office of the President Johns Hopkins UniversityLainie Rutkow JD MPH PhD Senior Advisor to the President for National Cap-
ital Academic Strategy Office of the President Johns Hopkins UniversityJosh Sharfstein MD Vice Dean for Public Health Practice and Community
Engagement Professor of the Practice Johns Hopkins Bloomberg School of Public Health
Jeremy Sugarman MD MPH MA Deputy Director for Medicine Harvey M Meyerhoff Professor of Bioethics and Medicine Johns Hopkins Berman Institute of Bioethics Department of Medicine Johns Hopkins School of Medicine and Department of Health Policy and Management Johns Hop-kins Bloomberg School of Public Health
Eric Toner MD Senior Scholar Johns Hopkins Center for Health Security Senior Scientist Department of Environmental Health and Engineering Johns Hop-kins Bloomberg School of Public Health
Marc Trotochaud MSPH Analyst Johns Hopkins Center for Health Security Research Associate Department of Environmental Health and Engineering Johns Hopkins Bloomberg School of Public Health
Effy Vayena PhD Professor Health Ethics amp Policy Lab Department of Health Sciences amp Technology ETH Zurich
Tal Zarsky JSD LLM LLB Professor of Law University of Haifa Faculty of Law Visiting Scholar University of Pennsylvania Law School (2019ndash2020)
Preface
ix
Digital technologies are being developed and promoted to support the public health response to the COVID-19 pandemic with discussion and implementation planning in the United States by localities states institu-tions and employers Key decision makers and stakeholdersmdashincluding government officials institutional leaders employers digital technology developers and the publicmdashrequire clear and well-supported guidance to inform the deployment and use of these technologies as well as of the data they collect store and share While technology-based approaches are currently unable to provide solutions on their own experiences in other countries indicate that they could be used successfully in conjunc-tion with traditional and novel public health methods
This report reflects a rapid research and expert consensus group ef-fort led by the Berman Institute of Bioethics and the Center for Health Se-curity at Johns Hopkins University It draws on experts from both inside and outside Johns Hopkins in bioethics health security public health technology development engineering public policy and law The report highlights issues that must be addressed and provides recommendations for the use of digital technologies as part of contact tracing
The analysis offered here is focused on answering the following questions
bull Can digital contact tracing technologies (DCTT) be effective as part of public health responses to the pandemic and if so to what degree for which specific types of functions with what confidence and with what requirements
bull How can these technologies serve the interests of public health while respecting other individual and collective interests such as ensuring equitable distribution of benefits and burdens and limit-ing infringement on privacy and other civil liberties
x Preface
bull What are the ethical legal policy and governance guardrails cur-rently in place around such technologies and what else is needed
bull What additional guardrails are required to ensure that the goals of public health in using these technologies are achievable in ways that are ethically and legally sound
To answer these questions the report examines some core aspects of dig-ital technologies applied to contact tracing focusing on
bull the value of and basic methods for traditional public health surveil-lance and contact tracing
bull candidate technological products to enhance public health surveil-lance and contact tracing how they work and their comparative value for public health
bull core ethical legal and governance considerations and how they relate to relevant features of candidate technological solutions and
bull what is needed to move forward responsibly with the use of digital technology in support of public health surveillance acknowledg-ing gaps in our current understanding
The project involved in-depth analysis by a dedicated team of faculty postdoctoral fellows and research staff working over the course of only a few weeks but with great intensity drafting a report in collaboration with 26 total contributors writing commenting and revising through multiple drafts with the penultimate draft ldquopressure-testedrdquo by review and discus-sion at a virtual workshop of invited experts and stakeholders held on May 13 2020 and the final version completed on May 21 2020 The report and analysis builds on the excellent work of others in some parts of this territory while focusing on the gaps in analysis and areas that have not been sufficiently addressed The goal is to offer comprehensive guid-ance to relevant stakeholders to advance public health response during the COVID-19 pandemic Given the rapidly evolving territory into which DCTT is being introduced this report will by necessity be something of a living document updated as often as information dictates in order to continue to offer leading-edge analysis and guidance Versions will be noted in the digital and print editions
Acknowledgments
xi
Efforts like this project require teams and even small armies to be carried out successfully and this was no exception except that it was many fewer people working many more hours than could reasonably be expected of them From the initial kernel of an idea to the publication of this report in book form this project took just over a month total That seems impossi-ble even as I know it is accurate and it speaks to the incredible commit-ment hard work research skills and analytic acumen of our colleagues at Johns Hopkinsmdashthe core team are deservedly listed as lead authors of this report
None of this would have been possible without the supportmdashmoral and financialmdashand encouragement of Johns Hopkins University Presi-dent Ronald J Daniels who was the first to suggest the idea to me of taking on this topic He provided not only support and encouragement but the imprimatur of his office including help guidance and counsel from Prof Lainie Rutkow senior advisor to the president Lainie played a more integral role than that description captures reflected in part by her inclusion among the reportrsquos contributors but she deserves special acknowledgement for shepherding us through to the end
I mentioned that this was a team effort and every team requires an effective leader My colleague Prof Joseph Ali stepped into that role as we undertook the project and then he worked seven days a week along with the rest of the core research and writing team always unfailingly positive and deeply engaged in the work He along with Prof Anne Barnhill Alan Regenberg Amelia Hood and Drs Katelyn Esmonde Brian Hutler and Crystal Watson all deserve special thanks for doing so much in so little time all while working under the grinding social distancing restrictions of the 2020 pandemic That work was supported by Arnold amp Porter Kaye Scholer LLP with legal research and other assistancemdasha huge thanks to
xii Acknowledgments
them Finally the 16 contributing authors were incredibly generous with their time energy and insights all on ridiculously tight timelines and never a complaint or objection
The project benefited greatly from a number of experts who provided written feedback on drafts and who attended the virtual workshop to test our recommendations including Miles Stewart Rob Nichols Smisha Aagarwal Karl Steiner Anupam Joshi Charles Scheeler Ford Rowan and Jay Wagley
Last the fact that this report appears in published book form by Johns Hopkins University Press is another minor miracle from manu-script to printed book in under a week Thanks to JHUP Director Barbara Kline Pope and her team for being willing to take on the challenge and for the incredible focused effort it required
my heartfelt thanks andappreciation to you all
Jeffrey Kahn
Acronyms and Abbreviations
xiii
ADA Americans with Disabilities ActBLE Bluetooth Low EnergyCalOPPA California Online Privacy Protection ActCBP Customs and Border ProtectionCCPA California Consumer Privacy ActCDC Centers for Disease Control and PreventionCLOUD Act Clarifying Lawful Overseas Use of Data ActCOPPA Childrenrsquos Online Privacy Protection ActCOV+ confirmed positive SARS-CoV-2 test resultCOVID-19 coronavirus disease 2019CPNI customer proprietary network informationCSLI cell-site location informationDCTT digital contact tracing technology and closely related
digital health productsECPA Electronic Communications Privacy ActEEOC Equal Employment Opportunity CommissionE-SIGN Electronic Signatures in Global and National
Commerce ActEU European UnionFCC Federal Communications CommissionFTC Federal Trade CommissionFTCA Federal Trade Commission ActGIS geographic information systemGPS global positioning systemHHS US Department of Health and Human ServicesHIPAA Health Insurance Portability and Accountability ActHIV human immunodeficiency virusICU intensive care unit
xiv Acronyms and Abbreviations
IRB institutional review boardJHU Johns Hopkins UniversityOCR Office for Civil Rights US Department of Health amp
Human ServicesOSHA Occupational Safety and Health AdministrationPHI protected health informationPII personally identifiable informationPPE personal protective equipmentPPPT privacy-preserving proximity trackingQR Code quick response codeRFRA Religious Freedom Restoration ActRLUIPA Religious Land Use and Institutionalized Persons ActSARS-CoV-2 severe acute respiratory syndrome coronavirus 2SCA Stored Communications ActSTI sexually transmitted infection
DIGITAL CONTACT TRACING FOR PANDEMIC RESPONSE
11
Summary
Introduction
Public health professionals around the world are working tirelessly to respond to the COVID-19 pandemic using tried-and-tested public health methods for infectious disease surveillance and control These traditional methods are essential to the global COVID-19 response To complement these actions and potentially augment the speed and efficacy of the pub-lic health workforce digital technologies are being harnessed Given the scale of the pandemic significant efforts are being undertaken to develop and leverage public-facing and health-system-supportive technology solu-tions including smartphone apps and other digital tools that may aid public health surveillance and contact tracing
Digital contact tracing technology and closely related digital health products (together DCTT) have been used in several countries as part of broader disease surveillance and containment strategies In the United States DCTT has been proposed as an integral part of some plans to ldquoreopenrdquo the country (Allen et al 2020 Hart et al 2020 Simpson and Conner 2020) It is almost certain that these and related technologies will become part of not only the COVID-19 response but also the larger toolbox for future public health communicable disease prevention and control
These technologies have significant promise They also raise import-ant ethical legal and governance challenges that require comprehensive analysis in order to support decision-making Government officials pub-lic health leaders leaders of institutions employers digital technology de-velopers and the public all must be adequately informed in order to make
2 Digital Contact Tracing for Pandemic Response
responsible choices Johns Hopkins University recognized the importance of helping to guide this process It organized an expert group with mem-bers from inside and outside of Hopkins and led by its Berman Institute of Bioethics in collaboration with the Center for Health Security Its charge was to examine the ethics law policy and public health implications of using digital technologies as part of pandemic response and to develop guidance including a framework and actionable recommendations for governmental and institutional decision makers
Overall this expert group urges a stepwise approach that prioritizes align-ment of technology with public health needs and public values building choice into design architecture and capturing real-world results and impacts to allow adjustments as required Further we urge an approach that recog-nizes that there are complicated issues to resolve for governments insti-tutions and businesses and that introduction of DCTT must include public engagement and ongoing assessments to improve both performance and adoption
Specific recommendations include the following
bull There is no ldquoone size fits allrdquo approach to DCTT Technology design should not be static but it should be capable of evolving depending upon local conditions new evidence and changing preferences and priorities
bull Technology companies alone should not control the terms condi-tions or capabilities of DCTT nor should they presume to know what is acceptable to members of the public
bull DCTT should be designed to have a base set of features that pro-tect privacy with layers of additional capabilities that users may choose to activate An initial default should be that user location data are not shared but users should be provided with easy mech-anisms and prompts to allow for opting-in to this capability with encouragement to the public if it is shown to be critical to achieving public health goals
bull Data collected through DCTT should be made available to public health professionals and to researchers in de-identified form to support population-level epidemiologic analyses
Summary 3
bull Those who authorize use of DCTT within a particular jurisdic-tion or institution should continuously and systematically monitor the technologyrsquos performance in that context This should include monitoring for effectiveness and benefit monitoring for harms and monitoring for the fair distribution of both benefits and harms
bull Governments should not require mandatory use of DCTT given uncertainty about potential burdens and benefits Additional tech-nology user and real-world testing is needed
Through in-depth analysis and recommendations this report seeks to guide decision-making and enhance understanding of
bull the value of and basic methods for traditional public health surveil-lance and contact tracing
bull candidate technological products to enhance public health surveil-lance and contact tracing and their comparative value for public health
bull core ethical legal policy and governance considerations and how they relate to relevant features of candidate technological solutions and
bull what is needed to move forward responsibly with the use of digital technology in support of public health surveillance acknowledg-ing gaps in our current understanding
The full set of recommendations are intended to (1) support effective and informed adoption of DCTT (2) encourage design of flexible technol-ogies that maximize public health utility while respecting other values (3) establish meaningful processes for user disclosure and authorization (consent) (4) promote equity and fairness in the uses of DCTT and (5) foster transparent governance and oversight
DCTT Features Functions and Potential Applications
Digital contact tracing technologies and platforms can be roughly catego-rized into three broad approaches along a spectrum of potential policies and methods a maximal approach (typified by the South Korean govern-
4 Digital Contact Tracing for Pandemic Response
mentrsquos centralized and triangulated data collection (M S Kim 2020)) a minimal approach (typified by the AppleGoogle decentralized priva-cy-preserving proximity tracking (PPPT) and contact notification (Apple and Google nd)) and a diverse range of middle-ground approaches that aim to augment manual contact tracing with the collection of digital data that can be shared with public health authorities
Minimal approaches such as the AppleGoogle PPPT use Bluetooth Low Energy (BLE) ldquohandshakesrdquo that record close contact between mo-bile phone users but do not register the location in which the contact hap-pened In most architectures these proximity data are stored in the usersrsquo phones as anonymized ldquobeaconsrdquo that cannot be used to re-identify the users directly If a user with a PPPT app installed on their phone tests pos-itive and enters test results into their app those who have been in contact with them can be notified by the app This ldquoexposure notificationrdquo can be automatic or at the discretion of the COV+ person depending on the app design If notified a user who has been in contact with a COV+ individ-ual would receive a push notification alerting them to possible exposure (which may be timestamped) but with no other identifying information
The most prevalent middle-ground approach in the US context in-volves the collection and storage of personal datamdashincluding identifying information and location datamdashon the userrsquos phone These decentralized but personally identifiable data can then be voluntarily shared with pub-lic health officials if the user tests positive for SARS-CoV-2 (severe acute respiratory syndrome coronavirus 2) For example a team at the Mas-sachusetts Institute of Technology (MIT) has developed an app called Private Kit Safe Paths (MIT nd) that stores usersrsquo location data on their phone for 28 days If a user tests positive she can voluntarily upload her location data to a website that is accessible only to public health officials Officials can then analyze these personally identifiable data and subse-quently broadcast redacted and de-identified data to other users Healthy users would have access to these redacted location data of COV+ users but their own data would not leave their phones At a minimum the storage of user location data can function as a ldquomemory aidrdquo if the user tests positive but releasing the data to public health authorities may help to analyze the spread of SARS-CoV-2 and alert individuals or groups that have been in contact with COV+ patients
Summary 5
The US Centers for Disease Control and Prevention (CDC) has pub-lished preliminary criteria for evaluating capabilities and attributes of DCTT (CDC 2020e) These and other resources suggest that a compre-hensive assessment of DCTT and its potential to advance the publicrsquos health will require careful consideration of numerous interconnected fac-tors that interact in complex ways and must be navigated within the chal-lenging contexts of uncertainty and urgent need (Figure 1) These include
bull scientific and epidemiological understanding of SARS-CoV-2 transmission and infection
bull public health needs for combating the outbreak
bull technological capabilities of DCTT
bull performance of DCTT applications
bull ethical values and principles
bull characteristics of public adoption and acceptance and
bull legal issues and landscape
FIGURE 1 Interrelating Factors That Frame Responsible Development of Digital Contact
Tracing Technology
6 Digital Contact Tracing for Pandemic Response
The primary objectives for use of DCTT during the COVID-19 pandemic must be to reduce illness and death and facilitate public health efforts to reduce transmission of the virus These objectives fall under a broader overall goal of contributing to societal well-being during the pandemic It is not yet known whether and how much DCTT can contribute to these primary objectives nor whether it will be able to contribute without gen-erating new burdens or even harms such as incorrect warnings or ldquonoiserdquo that detract from the work of manual contact tracing
The process of identifying acceptable technology designs and uses is complex given the interplay among the factors Our analysis reveals that there is no ldquoone size fits allrdquo approach to DCTT There is variability across the United States with respect to SARS-CoV-2 prevalence and in-fection rates public health capacity public attitudes toward DCTT and acceptability of various potential features Moreover our understanding of SARS-CoV-2 and DCTT is evolving public health response needs and capabilities are changing and public attitudes are shifting Different tech-nologies used in different ways may be appropriate to achieve slightly dif-ferent public health goals in different localities and at different points in the pandemic A tiered and phased approach to technology development should be facilitated by law and policy prioritizing underlying interoper-ability while permitting user choices now and for the future
Given the complexity of the terrain as a first step those developing or considering widespread use of DCTT as part of pandemic response should be guided by the following principles and related actions (see box) These principles are meant to apply to DCTT as well as other dig-ital technologies used in novel ways during pandemic response
These principles make clear that in order to maximize the public good from use of DCTT public health needs and technological capabili-ties must be carefully aligned Government officials public health leaders leaders of other institutions employers digital technology developers and the public are all key stakeholders that must be informed and en-gaged in order to enable the most successful and ethically acceptable uses of DCTT
Guiding Principles for the Use of Digital Public Health Technologies for Pandemic Response
Transparency and public engagement are essential to an inclusive digital public health response
bull Government public health and digital technology leaders must engage
effectively with the public and other stakeholders to communicate the util-
ity importance oversight and limitations of relevant digital technologies
including their implications for individualsrsquo privacy and civil liberties
bull Transparency at all levels is essential for maintaining public trust and
confidence
bull To the extent possible digital public health responses should reflect the
range of values that are important to individuals including advancing the
health and well-being of the community as a whole
bull Decision makers should recognize the sacrifices that some people may be
willing to make during a pandemic in order to advance public health goals
Acceptance by some of particular monitoring capabilities should not be read
as a willingness to extend these methods to other problems or uses
Digital public health responses must represent the least infringement of civil liberties necessary to accomplish the public health goals
bull If preferred digital public health strategies infringe on privacy and other civil
liberties the infringements must be sufficiently justified by the circum-
stances of the pandemic offset by ample anticipated public benefit and
considered relative to infringements associated with other possible strate-
gies such as mass physical distancing
bull Only those data that are necessary and relevant for the stated public health
purposes should be collected Identifiable data should be stored in a secure
manner and only for the period of time that the public health purposes
require
bull Adopted technologies should not be used in ways that subject communities
to discrimination or surveillance for nonndashpublic health reasons
bull Respect for individual autonomy requires that users are sufficiently informed
of the public health goals of the technology and the extent to which those
goals are being met
Use of digital public health technologies and data must be guided by best available evidence
bull Decisions to deploy digital public health technologies should be based on a
careful assessment of the uses and limitations of any proposed technology
taking into account the best available evidence
bull Those who deploy digital public health technologies should continuously and
systematically monitor their performance as well as any evidence that is being
generated in other contexts about the selected technological solution and about
other competing technologies
bull Unintended consequencesmdashincluding those that might impact public health
goals core values and interests of the public and unfair advantage or disadvan-
tagemdashshould be carefully monitored and addressed as necessary
Responsible use of digital public health technology requires meaningful governance and accountability
bull Systems of governance must be trustworthy and well informed They must be
reviewed and adjusted as circumstances and evidence change or as unintended
effects are identified
bull Trusted representatives who are capable of developing and implementing uni-
form and fair standards for adopting and utilizing underlying digital technology
must be identified
bull Understandable transparent and publicly accessible rules must guide the
collection access control use storage and combination of data by government
authorities public and private institutions and other parties such as public
health researchers
bull Oversight accountability and consequences for abuse or misuse of these data
must be explicit and enforceable
The deployment of digital public health technology must be rooted in a commitment to equity
bull Digital public health technologies should be deployed in a manner that does
not propagate preexisting patterns of unfair disadvantage or further distribute
harms and risks unfairly throughout the population
bull To the extent possible digital public health technologies should be designed to
rectify existing inequities
bull Oversight mechanisms must be in place to ensure that the improved public
health outcomes are equitable and to detect and correct any unforeseen resul-
tant injustices attributable to the technology or that can be addressed using the
technology
bull The incentives and disincentives for adopting new technology must be equita-
ble not exploitative and aligned with effective use of the technology
bull Disparity-driven technology gaps should be explicitly recognized To the extent
possible provisions should be made to address the digital divide
Summary 9
Summary of Recommendations
The guidance document makes a number of recommendations related to (1) supporting effective and informed adoption of DCTT (2) designing flexible technologies to maximize public health utility while respecting other values (3) establishing meaningful processes for user disclosure and authorizationconsent (4) promoting equity and fairness in application of DCTT and (5) instituting transparent governance and oversight Here we provide a summary of recommendations
Supporting Effective and Informed Adoption
bull Those who authorize use of DCTT within a particular jurisdic-tion or institution should continuously and systematically monitor the technologyrsquos performance in that context This should include monitoring for effectiveness and benefit monitoring for harms and monitoring for the fair distribution of both benefits and harms They should also monitor evidence that is being generated in other contexts about their selected technological solution and about other competing technologies
bull Data collected through DCTT should be made available to public health professionals and to researchers in de-identified form to support population-level epidemiologic analysis
bull Data should be available to users that would permit them to further investigate their personal risk with public health officials or other health workers to add a layer of protection against unnecessary quarantine
bull Technologies or apps may produce some false negatives or false positives but they should be accurate enough that public health authorities feel confident that they support and donrsquot detract from contact tracing efforts
bull Trusted leaders should be enlisted to communicate effectively with the public about DCTT and encourage its use should the technol-ogy demonstrate some potential The limits of knowledge regard-ing effectiveness should also be explained along with what will be done to improve technological capabilities as understanding evolves
10 Digital Contact Tracing for Pandemic Response
bull Incentives can be a useful complement to encouragements how-ever any incentives for users to install and use DCTT must be equitable should not be coercive and should align with effective use of the technology
bull DCTT use should not be mandated at this time given uncertainty about potential harms and benefits Additional technology user and real-world testing is needed
Designing Flexible Technology to Maximize
Public Health Utility While Respecting Other Values
bull Technology companies should not alone control the terms condi-tions or capabilities of DCTT nor should they presume to know what is acceptable to members of the public
bull A ldquovalues in designrdquo approach to development of DCTT should be adopted (Flanagan Howe and Nissenbaum 2008 Knobel and Bowker 2011) Robust public and user engagement activities should be pursued to identify and incorporate to the extent pos-sible a range of values into the design of the technology These values may include privacy but also autonomy efficiency equity or others Technology design should reflect an appropriate balance and prioritization of identified values
bull Technology design should not be static but should be capable of evolving depending upon local conditions new evidence and changing preferences and priorities
bull DCTT should be designed to have a base set of features that pro-tect privacy with layers of additional capabilities that users may choose to activate An initial default should be that user location data are not shared but users should be provided with easy mech-anisms and prompts to allow for opting-in to this capability with encouragement to the public if it is shown to be critical to achieving public health goals
Establishing Meaningful Processes for
User Disclosure and Authorization (Consent)
bull A clear and concise module consisting of basic disclosure and vol-untary authorization should be developed to accompany DCTT
Summary 11
This module should not take the form of ldquoclickwraprdquo terms of service or end-user agreements but rather provide only essential information necessary for an individual to make a decision More detailed disclosures (such as FAQs in plain language) should be made easily accessible to those who wish to learn more with no hidden surprises
bull An opt-in approach to authorization should be instituted to ac-company initial DCTT rollout The feasibility and value of opt-out approaches should continue to be evaluated informed by what is technologically possible what local assessments of benefits and harms of the technology reveal over time and our evolving un-derstanding of the degree to which an opt-out approach is likely to increase or decrease utilization among different populations Opt-out approaches should not be precluded
Promoting Equity and Fairness in Application of DCTT
bull States localities and institutions that recommend widespread use of DCTT should provide technology (eg mobile phones Blue-tooth devices) and free data packages to those who desire but lack access to these devices
bull If there are lower rates of adoption of DCTT systems in some iden-tifiable communities public health authorities should find ways to compensate For example directing more non-DCTT resources and efforts toward those communities to meet specific needs that are elsewhere being supported by technology
bull If maps are generated based on DCTT to provide the public with the locations that COV+ individuals have visited steps must be taken to minimize the stigma and potential financial losses that could result from a location being identified as a hotspot
Instituting Transparent Governance and Oversight
bull Digital surveillance oversight committees should be established expeditiously with diverse and qualified membership to provide ethical and regulatory review prior to and concurrent with wide-spread use of a DCTT system
bull Understandable and publicly accessible rules must guide the col-
12 Digital Contact Tracing for Pandemic Response
lection access control use storage and combination of data by government authorities public and private institutions and other parties such as public health researchers
bull Only those data that are necessary and relevant for the public health response to COVID-19 should be collected and used
bull Identifiable data should be kept only for the period of time needed for the public health response to COVID-19
bull Identifiable data collected as part of this response should not be shared with anyone other than the relevant public health authorities without additional specific informed consent of individual users
bull Before a government or institution adopts a digital contact tracing program they should state the conditions under which the digital contact tracing program will be terminated
bull Future use of DCTT to advance public health or other efforts (eg use in seasonal flu surveillance) would require independent justification DCTT designed for public health use should not be used by law or immigration enforcement
bull The principles offered in this guidance document apply both during and following the COVID-19 pandemic
Legislative Recommendations
bull The United States Congress should enact new legislation specif-ically tailored to facilitate the use of DCTT as part of the public health response to COVID-19 while also protecting user privacy and ensuring data security
bull Congress should require DCTT developers to disclose to users in clear language the nature of the information that would be col-lected how it would be collected how it would be stored and for what purposes it may be used
bull While the rollout of DCTT should initially employ an opt-in au-thorization approach the feasibility acceptability and value of opt-out approaches should continue to be evaluated As such opt-out approaches to consent should not be precluded by legislation
Summary 13
bull Congress should prohibit the commercial use of data collected for COVID-19 response by DCTT
bull Congress should prohibit discrimination on the basis of data col-lected by DCTT
bull If Congress is unable to enact suitable legislation state legislatures should work toward enacting similar laws for their jurisdictions A ldquomodelrdquo state law should be rapidly developed to facilitate na-tionwide adoption of an appropriate law and uniformity of legal requirements
Summary of Analysis
Supporting Effective and Informed Adoption
The COVID-19 pandemic and the physical distancing efforts imple-mented to slow the rate of transmission have caused severe harm to indi-viduals communities and our society To protect the public good going forward we need a robust public health response that reduces the spread of SARS-CoV-2 and does so in a way that allows economic recovery to occur and to be sustained We also need to design and manage this public health response so as to minimize harms to individuals and society to distribute benefits and burdens equitably across the population and to avoid misuses of the technologies and the data they collect
To reduce the spread of SARS-CoV-2 chains of transmission need to be broken To do this people who have been exposed to SARS-CoV-2 or potentially exposed need to be identified as comprehensively and as quickly as possible so they can quarantine themselves and avoid infecting others This is the job of manual contact tracing by public health authori-ties in which people infected or presumptively infected with SARS-CoV-2 are interviewed and asked about their movements and interactions in-cluding where they work and shop how they travel with whom theyrsquove had contact and the nature of that contact (eg where the contact took place) Their contacts are then interviewed and potentially asked to quar-antine seek testing and take other protective measures if the contact is sufficiently high risk
14 Digital Contact Tracing for Pandemic Response
The hope is that DCTT can augment traditional contact tracing ef-forts either by working alongside and independently of manual contact tracing or by being integrated into manual contact tracing efforts in a way that makes these efforts faster more thorough and more efficient
Data suggest that a substantial proportion of transmissionsmdashper-haps as high as 50mdashoccur between individuals who are not symptom-atic and that transmission may occur as early as 3 days before onset of symptoms (WHO 2020) Because asymptomatic spread of SARS-CoV-2 appears to be a significant source of infection we need to identify po-tentially infected people before they show symptoms thus speed is of the essence This is one benefit of using DCTT potential contacts can be identified instantaneously notified quickly and asked to quarantine as soon as possible
Another benefit is identifying contacts who manual contact tracing methods may miss either because COV+ people do not remember all the places theyrsquove been or cannot identify all the people theyrsquove had contact with This is especially relevant given the long period of infectivity of SARS-CoV-2 which begins before people are symptomatic and aware they are infected (Ferretti et al 2020) If DCTT were designed to have optional location-monitoring capabilities this critical challenge could be mitigated even further For example location data might reveal that a COV+ person was at a restaurant at an exact time and date which could be followed up by contact tracers who could alert the public or use other measures to reach those who were also present in the restaurant at the same time In other disease contexts geolocation data have demonstrated some potential to support epidemiology and disease surveillance (see Fur-lanello et al 2002 Dredze et al 2013 Eckhoff and Tatem 2015 Fraccaro et al 2019) with technical cautions regarding accuracy and the like (Beu-kenhorst et al 2017)
One role for DCTT is to work alongside manual contact tracing but independently of it Individuals would download proximity tracing or exposure notification apps use them receive alerts if theyrsquove had a poten-tial contact with another user who is COV+ or presumptively COV+ and voluntarily self-quarantine without having contact with public health au-thorities or giving them data that feeds into public health contact tracing efforts It is possible that this would help to break chains of transmission and reduce the spread of SARS-CoV-2 though at this point these benefits
Summary 15
are speculative It is also possible that such exposure notifications will result in high rates of false positives
Another possible role is for DCTT to be integrated into manual con-tact tracing efforts When potential contacts are identified by DCTT they are connected to public health authorities who can then follow up with them There are different forms this could take and different kinds and amounts of data about contacts public health authorities could receive from DCTT On one end of the spectrum of reporting public health authorities would not receive individualsrsquo names or contact information only anonymous data The fullest version of reporting would securely send to public health authorities the names contact information such as address and phone number and other data about contacts that DCTT collected including data about their location and movement history
It is uncertain whether providing public health authorities with vol-umes of information on cases and contacts from DCTT will be useful in practice As mentioned above providing public health authorities with location data on cases and contacts collected by DCTT may help con-tact tracers to find and notify additional contacts However at present providing public health authorities with large amounts of data will be useful only if there is sufficient capacity to follow up on these data In addition there is a risk of low-quality data from DCTT flooding the system leading to investigation of false case contacts identified by DCTT and distracting from other important efforts Whether and to what extent data from DCTT will benefit contact tracing efforts is unknown pointing again to the importance of continuously collecting high-quality evidence about DCTT
Designing Flexible Technology to Maximize
Public Health Utility While Respecting Other Values
Use of DCTT is essentially an experiment as we have insufficient infor-mation about the performance of different DCTT and their efficacy In the face of this uncertainty how should DCTT be designed and how should its use be managed
Many efforts to advance DCTT in the United States and elsewhere have emphasized the importance of ldquoprivacy by designrdquo that is building privacy and security protections into the design of technology rather than counting on responsible use alone (Cavoukian 2010) As noted above
16 Digital Contact Tracing for Pandemic Response
some major technology companies have signaled this position through de-velopment of PPPT systems that embed features such as decentralization de-identified information user anonymity bans on collection of location data and minimal reliance on or integration of public health authorities or other government actors Many of these features have also been em-braced early by advocacy organizations (Crocker Opsahl and Cyphers 2020 Electronic Privacy Information Center 2020 Kahn Gilmor 2020) and in an open letter (ldquoJoint Statement on Contact Tracingrdquo 2020) from nearly 300 researchers These same actors have emphasized that use of DCTT should be fully voluntary
Although privacy is a key value individuals and communities may also value efficiency equity liberty autonomy economic well-being com-panionship patriotism or solidarity among other values People may accept more significant encroachments on privacy now if this ultimately results in realizing other values (such as companionship) that are of equal or greater importance to those individuals Rather than centering pri-vacy alone in design a different orientation is needed at this moment that of ldquovalues in designrdquo which incorporates a broader range of values into technology (Flanagan Howe and Nissenbaum 2008 Knobel and Bowker 2011) For example some users might wish to express autonomy solidarity or patriotism through DCTT by sharing their location history with public health professionals in order to advance the public health re-sponse increase system efficiencies (eg by contributing information that can lead to better data processing) and reduce the burden on essential workers At the same time there is value in further advancing autonomy by designing technology to allow individuals some control over what data about them are collected and shared
DCTT should be designed to have a base set of features that protect privacy and strive for interoperability but also should include other op-tional capabilities This could be achieved by designing DCTT to have a default that can be modified for example an initial setting could be that usersrsquo location data are not shared with public health authorities but us-ers may opt-in to this feature Such an opt-in approach is likely consistent with existing federal privacy laws
Designing DCTT this way gives users the flexibility to decide how to use the technology and how to engage with public health authori-ties consistent with their values and trade-offs they are willing to make This flexibility could also allow for more real-world evaluation of how
Summary 17
different users experience different features of DCTT in different loca-tions Technology design should not be static but it should be capable of evolving depending upon local conditions new evidence and changing preferences and priorities
DCTT developers must comply with a number of federal privacy laws These privacy laws generally permit the collection storage and use of personal information so long as the user provides meaningful consent Privacy law in the United States is generally sector-specific and limited in scope resulting in a patchwork of protections that differ significantly depending on the entity that collects the data and the type of data col-lected Given the complexity of existing federal privacy law and the need to further strengthen public trust in DCTT it would be beneficial for Congress to enact new privacy legislation that is specifically tailored to the use of DCTT in response to COVID-19 Such COVID-specific legisla-tion should be sensitive to the full range of values and recommendations described above
In short designing ldquomiddle-groundrdquo DCTT for flexible use may pro-vide the most adaptable and thus most robust public health responsemdashrespecting privacy and individual autonomy by allowing users to use DCTT in ways that express their own values
Public Acceptance of DCTT
While some groups have maintained that only PPPT-like minimal systems will be widely adopted because only they will earn and maintain public trust (Simpson and Conner 2020) there is insufficient evidence that pub-lic trust would be threatened by a DCTT system that has the capacity to securely collect location data integrate public health authorities and en-able voluntary sharing of certain user data (eg location data) with those authorities More research including through deliberative engagement sessions is needed to better understand how differences in the features and functionality of DCTT (such as optional sharing of geolocation data) influence trust and peoplersquos willingness to use DCTT Technology com-panies should not alone control the terms conditions and capabilities of DCTT nor should they presume to know what is acceptable to members of the public
Significant concerns have also been expressed by privacy advocates (Guariglia 2020) and in the popular press (Giglio 2020) about ldquosurveil-lance creeprdquomdashthat is a belief that state or corporate actors will use new
18 Digital Contact Tracing for Pandemic Response
surveillance technologies capacities and permissions well beyond the purposes for which they were initially justified to the public and beyond the time when they are useful for the COVID-19 pandemic Surveillance creep is a serious concern and should be carefully guarded against how-ever the possibility of surveillance creep is not a sufficient reason to limit development of DCTT to minimal systems Instead protections should be put in place to ensure that only those data that are necessary and relevant for the public health purposes at hand are collected and used and data should be kept only for the period of time needed for those pub-lic health purposes For this reason we would support COVID-specific legislation that would impose strict limits on the use of DCTT data for nonndashpublic health purposes
Finally the use of DCTT during the current pandemic should not set a precedent for future public health use (eg use in seasonal flu surveil-lance efforts) Future use would require independent justification Further use of DCTT in other contexts (eg by law enforcement or immigration enforcement) is presumptively unethical
Encouraging Adoption of DCTT
Researchers have estimated perhaps conservatively that DCTT use by 80 of smartphone owners or 56 of the population overall will be needed to suppress the epidemic (Hinch et al 2020) These estimates also highlight that some decrease in transmission would be realizable even with lower rates of technology adoption
In the United States many advocates and researchers have argued that use of DCTT must be fully voluntary However experience from other countries suggests that when use of a digital contact tracing app is voluntary only a minority of the population will download it Instead of making use fully voluntary and initiated by users there are ways that DCTT could be put into use without usersrsquo voluntary choice For exam-ple use of an app could be mandated as a precondition for returning to work or school or even further to control entry into a facility or trans-portation (such as airplanes) through scanning of a QR code to demon-strate personal exposure levels (Gan and Culver 2020)
While these approaches are hard to imagine in the United States some have argued that mandatory use of DCTT could be ethical If man-dates increase adoption of DCTT and improve the public health response
Summary 19
this would reduce the likelihood of lockdowns which are harmful and a severe limitation of individual liberty applied on a mass scale On the other hand mandated use of DCTT systems may not be effective People may not adhere to the mandate by simply leaving their phone at home Perhaps more important should the technology not deliver the hoped-for benefits having mandated the use of an unproven technology could result in a loss of public trust in the technology in the entity instituting the mandate and in the larger public health response potentially lead-ing to noncompliance with public health recommendations more broadly (Bernstein et al 2019)
Any decision maker considering mandatory use including govern-ment officials institutional leaders and employers must convincingly address a number of considerations Particularly important is the need to identify reliable evidence that the DCTT would be effective and to ensure that the burdens and benefits of use are equitable and justifiable At this time mandated use of DCTT by states or institutions is not jus-tifiable given uncertainty about potential harms and benefits Additional technology user and real-world testing is needed before mandatory use should be considered
As with any public health effort the amount of evidence that must be offered to illustrate that the intervention or program can achieve its aims and the degree to which people should be able to exercise choice in their participation should be in proportion to the anticipated bur-dens of the intervention or program For example the permissibility of mandating use of DCTT by the public depends on factors such as the sensitivity of the data that are collected the extent to which public health is integrated within the DCTT system and what actions are taken in response to confirmed virus exposure or being identified as COV+ (eg forced quarantine) The more burdens that are placed on individualsmdashfor example whether people are ordered into quarantine if they have been exposed to the virus or if there are limited social supports for those in quarantinemdashthe greater the demand should be on the performance of the DCTT system
Perhaps the most effective way to generate widespread US adoption of DCTT will be to offer incentives for its use in other contexts gener-ally speaking small incentives have been shown to lead to an increase in desired outcome (Singer and Ye 2013 Lee et al 2014) Given the impor-
20 Digital Contact Tracing for Pandemic Response
tance of widespread use modest incentives ought to be considered if and when there is sufficient evidence of the utility of DCTT so long as those incentives are not mandates in disguise Another ldquofirst linerdquo approach to increasing use of DCTT is for trusted community leaders public figures health care professionals and other respected individuals to communicate with the public and their communities about DCTT and to encourage its use through public engagement campaigns if and when the technology demonstrates sufficient potential
Establishing Meaningful Processes for
User Disclosure and Authorization (Consent)
Any effort to roll out DCTT should ensure that users have a meaning-ful opportunity to review and understand information about the specific technology and its uses Moreover given the importance of public trust and the current crisis of public trust in governments and technology com-panies handling private digital information there is a strong ethics argu-ment for requiring consent from individual users We recommend a care-fully crafted version of what is sometimes called simple consent which consists of basic disclosure and voluntary agreement or authorization This disclosure should include information about the purposes of the technology the userrsquos options for collecting and sharing data purposes for which data can be used and any known risks among other informa-tion This information should be presented in an accessible format on any DCTT app and more detailed disclosures should be readily accessible for those who wish to review them
Through an opt-in mechanism such as clicking a button to signal agreement users should be able to indicate their intention to use a DCTT The opt-in approach is consistent with mechanisms for agreement to use other downloaded applications An opt-in approach should be part of the initial introduction of DCTT given the novelty of the technology and its uses and the need to build trust and confidence in the system Successes of opt-out approaches in other areas suggest that the feasibility and value of an opt-out approach to DCTT should be carefully evaluated particularly in conjunction with assessment of whether public health goals are being met (Rithalia et al 2009) Such assessments should be informed by what is technologically possible by local data regarding benefits and harms of the technology and by evolving understanding of the degree to which
Summary 21
an opt-out approach is likely to increase or decrease utilization among different populations
Promoting Equity and Fairness in Application of DCTT
Digital contact tracing technology should be designed and used in ways that as far as possible promote an equitable distribution of benefits and burdens DCTT should be deployed in a manner that does not propagate preexisting patterns of unfair disadvantage or distribute harms and risks unfairly throughout the population It is well known that some commu-nities have lower rates of technology and data access and therefore may benefit less from use of DCTT unless steps are taken to address these digital disparities Additionally should use of DCTT be made a require-ment for entry into a workplace into a school or onto transportation then those who currently do not possess the required technology must not be unfairly burdened through lack of access In order to mitigate this states localities and institutions that recommend widespread use of DCTT should provide technology (eg mobile phones Bluetooth de-vices) and free data packages to members of the community who desire but lack access to these devices
Some populations may also experience greater harm and greater fear of harm from having their data collected For example some groups such as African Americans Hispanic Americans Muslim Americans and undocumented immigrants have more reasonable fear of their data being handed over to law or immigration enforcement and some groups have lower levels of trust in public health due to past injustices (Auxier et al 2019 CSM 2017 Pew Research Center 2017 Rodrigues et al 2018) This further substantiates the need to limit use of any data gathered by DCTT to its public health purpose
Instituting Transparent Governance and Oversight
DCTT must be developed with an eye toward both present and future implications We are rapidly gaining knowledge about SARS-CoV-2 and COVID-19 but still have essential gaps in our understanding In the United States public health responses including DCTT will generally be developed and coordinated by individual states regional consortia (Res-ton Sgueglia and Mossburg 2020) and associations Good governance in this context requires transparency and the creation of oversight bodies
22 Digital Contact Tracing for Pandemic Response
with the appropriate expertise and representation to allow nimble and effective responses while serving as trusted representatives
In order to address the range of ethics and governance concerns that relate to the design and use of DCTT we recommend that digital surveil-lance oversight committees be established perhaps at a state level and with a platform for national coordination These committees can provide ethics and regulatory review prior to and concurrent with widespread use of DCTT The committees should be composed of a diverse group of experts capable of evaluating the quality of a DCTT system locally including members of communities that experience higher rates of digital disparity
When assessing the design and use of digital contact tracing systems these committees (and the public more widely) should consider not only the risks and benefits accrued during the COVID-19 pandemic but also implications for the future How can we navigate safe use of these tech-nologies in a way that preserves public trust in them and enables the possibility of future beneficial use
As a start it should be emphasized that the principles offered in this and other guidance documents do not apply only during the pandemic Future efforts to advance DCTT capabilities during quieter times should make every effort to follow them
23
Introduction
Public health professionals around the world are working tirelessly to respond to the COVID-19 pandemic using tried-and-tested public health methods for infectious disease surveillance and control These traditional methods are essential to the global COVID-19 response To complement these actions and potentially augment the speed and efficacy of the public health workforce digital technologies are being harnessed Given the scale of the pandemic significant efforts are being undertaken to develop and leverage public-facing and health-system-supportive tech-nology solutions including smartphone apps and other digital tools that may aid public health surveillance and contact tracing
Digital contact tracing technology and closely related digital health products (hereafter DCTT) have been used in several countries as part of broader disease surveillance and containment strategies Globally many digital COVID-19 contact tracing strategies have already emerged in response to the pandemic This is not surprising given the ubiquity of mobile phones and other digital devices around the world (ldquoMobile Cellular Subscriptions (per 100 People)rdquo 2018) experiences developed during prior outbreaks and pandemics and the pre-COVID-19 momen-tum behind using digital technologies to support individual and health system capabilities (WHO 2017 Mathews et al 2019 Aiello Renson and Zivich 2020 Mahmood et al 2020) In the United States DCTT has been proposed as an integral part of some plans to ldquoreopenrdquo the country (Allen et al 2020 Hart et al 2020 Simpson and Conner 2020) It is al-most certain that these and related technologies will become part of not only the COVID-19 response but also the larger toolbox for future public health communicable disease prevention and control
24 Digital Contact Tracing for Pandemic Response
While novel public health surveillance technologies such as DCTT have theoretical promise their effectiveness is unclear These technologies also raise important ethical legal and governance challenges that require comprehensive analysis in order to support decision-making regarding their appropriate use A number of frameworks recommendations and analyses have emerged recently in an effort to chart potentially ldquosaferdquo pathways for use of public health disease surveillance technology Many in the United States such as the Electronic Frontier Foundation Elec-tronic Privacy Information Center American Civil Liberties Union and the Center for American Progress are proposing that digital public health surveillance technologies must embrace strict data privacy protections decentralized data storage a high degree of anonymity and voluntary adoption (Crocker Opsahl and Cyphers 2020 Electronic Privacy Infor-mation Center 2020 Kahn Gilmor 2020 Simpson and Conner 2020) Others have argued that technologies that seek to enhance public health response during a pandemic should more closely align with the needs of public health professionals and the evidence-based procedures they follow stating that interests in serving the publicrsquos health ought to weigh more heavily in the necessary balancing of stakeholder interests (de Jong et al 2019 Watson et al 2020) This view is in part based on a recogni-tion that during countless other outbreaks the public has benefited from traditional disease surveillance and contact tracing which are heavily re-liant on centralized data storage and when necessary the collection of identifiable information These traditional approaches are governed by ethics principles (PHLS 2002) ethics guidelines (WHO 2017) and laws (ASTHO 2012) and digital technologies represent a new tool to support them
While debates and recommendations about appropriate design and use of DCTT have focused intensely on minimizing important data-related risks a wider lens is needed to fully appreciate the many additional criti-cal questions that need attention This report begins to grapple with these questions which are critical to address in order to guide responsible use of DCTT Given the complexity of the terrain as a first step toward estab-lishing a foundation for responsible decision-making regarding potential use of DCTT we offer a set of guiding principles (see box) These prin-ciples are meant to apply to DCTT as well as other digital technologies used in novel ways during pandemic response
Guiding Principles for the Use of Digital Public Health Technologies for Pandemic Response
Transparency and public engagement are essential to an inclusive digital public health response
bull Government public health and digital technology leaders must engage
effectively with the public and other stakeholders to communicate the util-
ity importance oversight and limitations of relevant digital technologies
including their implications for individualsrsquo privacy and civil liberties
bull Transparency at all levels is essential for maintaining public trust and
confidence
bull To the extent possible digital public health responses should reflect the
range of values that are important to individuals including advancing the
health and well-being of the community as a whole
bull Decision makers should recognize the sacrifices that some people may be
willing to make during a pandemic in order to advance public health goals
Acceptance by some of particular monitoring capabilities should not be read
as a willingness to extend these methods to other problems or uses
Digital public health responses must represent the least infringement of civil liberties necessary to accomplish the public health goals
bull If preferred digital public health strategies infringe on privacy and other civil
liberties the infringements must be sufficiently justified by the circum-
stances of the pandemic offset by ample anticipated public benefit and
considered relative to infringements associated with other possible strate-
gies such as mass physical distancing
bull Only those data that are necessary and relevant for the stated public health
purposes should be collected Identifiable data should be stored in a secure
manner and only for the period of time that the public health purposes
require
bull Adopted technologies should not be used in ways that subject communities
to discrimination or surveillance for nonndashpublic health reasons
bull Respect for individual autonomy requires that users are sufficiently informed
of the public health goals of the technology and the extent to which those
goals are being met
Use of digital public health technologies and data must be guided by best available evidence
bull Decisions to deploy digital public health technologies should be based on a
careful assessment of the uses and limitations of any proposed technology
taking into account the best available evidence
bull Those who deploy digital public health technologies should continuously and
systematically monitor their performance as well as any evidence that is being
generated in other contexts about the selected technological solution and about
other competing technologies
bull Unintended consequencesmdashincluding those that might impact public health
goals core values and interests of the public and unfair advantage or disadvan-
tagemdashshould be carefully monitored and addressed as necessary
Responsible use of digital public health technology requires meaningful governance and accountability
bull Systems of governance must be trustworthy and well informed They must be
reviewed and adjusted as circumstances and evidence change or as unintended
effects are identified
bull Trusted representatives who are capable of developing and implementing uni-
form and fair standards for adopting and utilizing underlying digital technology
must be identified
bull Understandable transparent and publicly accessible rules must guide the
collection access control use storage and combination of data by government
authorities public and private institutions and other parties such as public
health researchers
bull Oversight accountability and consequences for abuse or misuse of these data
must be explicit and enforceable
The deployment of digital public health technology must be rooted in a commitment to equity
bull Digital public health technologies should be deployed in a manner that does
not propagate preexisting patterns of unfair disadvantage or further distribute
harms and risks unfairly throughout the population
bull To the extent possible digital public health technologies should be designed to
rectify existing inequities
bull Oversight mechanisms must be in place to ensure that the improved public
health outcomes are equitable and to detect and correct any unforeseen resul-
tant injustices attributable to the technology or that can be addressed using the
technology
bull The incentives and disincentives for adopting new technology must be equita-
ble not exploitative and aligned with effective use of the technology
bull Disparity-driven technology gaps should be explicitly recognized To the extent
possible provisions should be made to address the digital divide
Introduction 27
In reflecting on these principles it becomes clear that if we wish not only to realize but to maximize the public good that might come from use of DCTT we must carefully define and responsibly align public health needs and capabilities with technological needs and capabilities We must understand that although technology may serve as a workforce multiplier it alone will not solve the public health challenges we face We must identify and address assumptions and misinformation about technologies and data use We must provide the means and opportu-nity for informed decision-making by the public and those who serve as our representatives Government officials public health leaders leaders of other institutions employers digital technology developers and the public all must be adequately informed and engaged in order to make the best decisions possible under the circumstances
ONE
29
Types of Information Collected through Contact Tracing
Data Collected from Infected Persons
Symptoms and Course of Illness
Information about COVID-19 patientsrsquo signs symptoms and course of illness is important to public health because it provides a basis for refining clinical case definitions and informing health care providers and the gen-eral public (CDC 2020c) This includes the specific signs and symptoms manifested by persons who are COV+ as well as the relative frequency and durations of different signs and symptoms This would also take into consideration those persons with no symptoms but who test positivemdashthose who are presymptomatic (develop symptoms later) those who are postsymptomatic (clinically recovered but still infectious) and those who never manifest illness at all
Typically contact tracing begins with a case in which a person has confirmation of infection by means of a diagnostic test However in some cases test results are not reported until several days later and individuals may be identified as ldquopresumptive positiverdquo cases until testing can be completed In these cases contact tracing efforts will need to be updated when test results are returned For example if a test comes back negative public health professionals will want to notify contacts that they no lon-ger need to quarantine
Public Health Perspective
30 Digital Contact Tracing for Pandemic Response
Movement and Contacts
In order to manage cases appropriately (identify and track the infected isolate the sick quarantine the exposed) public health officials need de-tails on each case (Resolve to Save Lives nd) First they need to know who and where the individual is That means personally identifiable infor-mation and contact information (address phone numbers email) It also means information about the nature intensity and duration of contact with individuals to whom they may have transmitted the disease This may include information about where the individual works and the kind of work they do (eg health care worker) how they travel (eg bus sub-way car) and where they shop or any other public venues they may have visited during a period of possible infectiousness (PIH 2020a) It may be helpful in certain circumstances for public health officials to ensure that suspected cases contacts or other high-risk individuals are following iso-lation and quarantine recommendations or orders
Contact tracing involves identifying all individuals who have had sig-nificant exposure to confirmed or probable cases during the time prior to and after the onset of symptoms both of which are times when the case is thought to be infectious (Africa CDC 2020) Contacts could be those who are caring for COVID-19 patients especially if they lacked proper PPE and those who had close interaction with the COV+ person over a sustained period of time particularly in enclosed spaces (PIH 2020a) For COVID-19 contacts are identified by asking a person with a confirmed or probable case about people they may have been within 6 feet of for 15 minutes or more starting from 48 hours before the onset of symptoms and lasting until the person is isolated (CDC 2020b)
Data Collected from Contacts of Infected Persons
Contact Details
In addition to the data collected from individuals with COVID-19 con-tact tracers will collect data from potentially exposed individuals (con-tacts) Information about the nature intensity and duration of contact with an infected person may be collected for a contact if information about the case is known to the contact These details can help a contact tracer more accurately determine whether the contact is at high or low risk for SARS-CoV-2 transmission and help determine whether a con-tact should quarantine for 14 days (the upper bound of the SARS-CoV-2
Public Health Perspective 31
incubation period) In addition public health professionals may gather contactsrsquo demographic information and other personal data to contribute to population-level disease surveillance and situational awareness about an epidemic (CDC 2005) However the information needed at baseline is only a personrsquos name and contact information
Symptoms (If They Develop) and Course of Illness as well as Information about Close Contacts
If a contact develops COVID-19 symptoms while in quarantine andor tests positive for the virus public health will then collect the data required for a COVID-19 case This includes collecting information on the con-tacts that a person may have had (if any) in the days immediately before and during the course of their infection
How Contact Tracing Information Informs Public Health Action
To reduce disease burden and help make ldquoreopeningrdquo safer during the COVID-19 pandemic the United States and other countries will need to identify gather information about and safely isolate cases and quar-antine their contacts to reduce community transmission (Watson et al 2020) Gathering information about possible cases and their contacts en-ables public health to break chains of transmission
Contact tracing involves stages (CDC 2020a) including
1 identifying an infected person as a COVID-19 case
2 identifying the close contacts of that case (Africa CDC 2020)
3 getting in touch with contacts
4 asking contacts to quarantine at home for 14 days
5 assessing contacts for possible symptoms and
6 following up with COV+ persons and their contacts to identify new or worsening symptoms and connect them with medical care if needed
Contact tracers also play an important role in providing resources for COV+ persons who are in home isolation and their contacts who are
32 Digital Contact Tracing for Pandemic Response
in home quarantine Knowing who and where cases and contacts are can enable provision of supplies such as digital thermometers or masks Effective contact tracing that enables isolated cases and quarantined con-tacts to remain at home also requires providing a range of social sup-port services or ldquocare packagesrdquo from delivering food and medicines to trash pickup Furthermore vulnerable individuals who are homeless or otherwise unable to sufficiently isolate or quarantine in their current living conditions may need to have alternative housing arranged to safely remain separated from others (CDC 2020b)
Finally contact tracers explain what quarantined contacts should do if they begin to develop symptoms consistent with COVID-19 (Africa CDC 2020) Depending on the context contact tracers may engage in active monitoring by regularly communicating with contacts about their health status through phone text message or possibly mobile applica-tions In rare cases public health can make quarantine mandatory and may monitor a quarantined individual to ensure that they do not break quarantine Contact tracers may also facilitate access to health care by providing telemedicine resources or other information and support for accessing medical care
Characteristics That Make Data Useful to Public Health for Reducing Disease Transmission
Data Access
If digital contact tracing technology and closely related digital health products (together DCTT) are intended to support the public health ac-tions described above and directly amplify public health capacity to con-duct case identification and contact tracing then data collected through DCTT must be accessible to public health authorities Identifying infor-mation and location data for cases and contacts of cases are necessary for public health use so that contact tracers can do their work to uncover ongoing transmission and enable isolation and quarantine These data should also be durable meaning that public health can return to the data in order to interact with and support cases and contacts These data can also be useful at a population level if de-identified and aggregated by illuminating trends in community transmission and providing support for decisions about resource allocation
Public Health Perspective 33
Data Format
Data should be provided to public health authorities in a usable format that is compatible with public health systems and that has the granularity and specificity of personal information that is needed for use in contact tracing Without personal identifiers the data cannot be used by public health workers to undertake contact tracing Data should also contain information about the nature of a contact including the proximity of the contact and number of minutes that the person was in contact with an infected individual Location data can also help public health author-ities to conduct contact tracing particularly when contact occurred in a crowded area and involved people who donrsquot know one another Loca-tion data from a case can help public health professionals identify con-tacts even when those contacts themselves are not using a contact tracing app because the data shows contact tracers where to look for additional contacts
Data Accuracy
Data that identifies individuals as having sustained contact with a case must be as accurate as possible If criteria for being considered a contact are too restrictive it may result in missed contacts and sustained chains of disease transmission If criteria are too broad it may result in unneces-sary restriction of movement which could have significant personal and economic consequences
Timeliness of Data
Data from cases and contacts must be timely in order to enable case-based management that will help reduce community transmission For contact tracing to be effective infected individuals need to be isolated and their contacts identified and quarantined as quickly as possible Testing for SARS-CoV-2 can take time sometimes many days for a test result Especially because SARS-CoV-2 is transmissible during the pre-symptomatic period data on symptomatic individuals should be made available to public health officials even before a positive test is returned in order to enable identification and quarantine of contacts right away If this information is delayed until a test result is received it may be too late to identify and quarantine contacts because contacts (if infected) will already be contagious and may have spread the virus to others
34 Digital Contact Tracing for Pandemic Response
VolumeAvailability of Data
The more that individuals opt to share their information to support con-tact tracing the more effective contact tracing will be in breaking chains of viral transmission and controlling epidemics of COVID-19 The exact proportion of cases and contacts that need to be identified in order to avoid large surges of cases which overwhelm health care systems is un-certain but the goal is to identify all infected cases and all close contacts of each case (PIH 2020b)
Recommendations
bull Technologies or apps with the goal of enhancing public health ca-pacity to identify cases and trace contacts in order to control the spread of SARS-CoV-2 should be designed to match functionality with that goal
bull Technologies or apps may produce some false negatives or false positives but they should be accurate enough that public health authorities feel confident that they support and donrsquot detract from contact tracing efforts
bull DCTT approaches for public health should be designed to facilitate the following
deg identifying contacts including those who may not be easily found otherwise
deg finding and notifying contacts rapidly before they develop symptoms if infected
deg analyzing the nature of contact to determine whether contact is high medium or low risk and to support decisions about whether a contact should quarantine and
deg following up with cases and contacts so that public health can provide resources to support isolation and quarantine
bull Data collected through DCTT should be made available to public health professionals and to researchers in de-identified form to sup-port population-level epidemiologic analysis
T WO
35
Characteristics of SARS-CoV-2 Relevant to Candidate Digital Solutions
The SARS-CoV-2 virus has some unique transmission characteristics and clinical manifestations that can help guide use of digital contact trac-ing solutions Individuals infected with this virus may or may not show symptoms or may show a range of different and sometimes nonspecific symptoms Estimates regarding the percentage of individuals who are in-fected but never develop symptoms is highly uncertain ranging from 5 to 50 (Heneghan Brassey and Jefferson 2020) Data suggest that a substantial proportion of transmissionsmdashperhaps as high as 50mdashoccur between individuals who are not symptomatic and that transmissibility may extend out as long as 3 days before the onset of symptoms (WHO 2020)
The complexity of asymptomatic and presymptomatic transmission makes it more difficult to identify all cases of COVID-19 It also means that manual contact tracing is less effective because people are unlikely to remember all of their contacts during the long period of infectivity (Ferretti et al 2020) however it does not negate the need for contact tracing Identifying symptomatic cases will still help greatly with slowing the spread because their contacts can be asked to quarantine to prevent them from spreading the virus if they are indeed infected This means that whether they become symptomatic or not contacts will be quarantined and the chain of transmission will be broken If contact tracing can be implemented on a large enough scale perhaps with support from DCTT
Digital Technology and Contact Tracing
36 Digital Contact Tracing for Pandemic Response
eventually the virus could be managed at much lower levels of community transmission and large epidemics of unrecognized spread will not occur
The transmissibility of the virus when a person has no symptoms further suggests that effective solutions may require multimodal inter-ventions combining contact tracing with frequent rapid and ubiquitous testing and continued social distancing to varying extents (Cheng et al 2020)
Because of presymptomatic spread contact tracing efforts and dig-ital solutions to augment those efforts should support identification of contacts a person had 2 days before their symptoms and at least 3 days after the resolution of those symptoms (if the person continued to have contacts through that time period) (CDC 2020d) Additionally public health messages delivered by these technologies should urge contacts to quarantine for the full 14-day incubation period
Previously Existing Contact Tracing Technologies
Prior to this pandemic health agencies in high- medium- and low-income countries had begun to develop and use digital tools to augment the man-agement of infectious diseases including sexually transmitted infections (HIV chlamydia gonorrhea) and high-consequence epidemics (Ebola) (Danquah et al 2019)) However these have been primarily used to facil-itate case interviews partner notification (in the case of STIs) and record keeping as opposed to fully digitizing or automating the contact tracing process
It has been recently suggested that digital contact tracing could con-tribute to the management of the ongoing COVID-19 pandemic and the experiences of containing SARS-CoV-2 in countries such as China Sin-gapore and South Korea provide noteworthy examples However un-dertaking this case-based intervention on the scale required to achieve pandemic control is a novelty in the history of public health Although technological development is proceeding rapidly several foundational is-sues have yet to be resolved including functionality connectivity to pub-lic health authorities and informatics systems usability by disease inter-vention specialists (DIS also referred to as contact tracers) and sufficient protection of personally identifiable information among others
Digital Technology and Contact Tracing 37
Introduction of Novel Digital Contact Tracing Technologies
Digital contact tracing technologies and platforms have recently been in-troduced and the CDC has published preliminary criteria for evaluating these tools (CDC 2020e) It can be helpful to consider three broad ap-proaches along the spectrum of potential methods of digital contact trac-ing a maximal approach (typified by the South Korean governmentrsquos cen-tralized data collection (M S Kim 2020)) a minimal approach (typified by the AppleGoogle decentralized privacy-protecting proximity tracking (Apple and Google nd)) and a diverse middle ground that aims to aug-ment manual contact tracing with the collection of digital data Perhaps the most promising approach in this middle ground involves allowing us-ers to turn over both proximity data and GPS location data (ie cell-site location data) to public health authorities on a voluntary basis
Along with this ldquominimal to maximalrdquo spectrum in the design of dig-ital contact tracing technologies and systems there is another spectrum that concerns voluntary versus mandatory use of these technologies are individuals entirely free to use these technologies or not or should poli-cies incentivize or even mandate their use At one extreme South Korea (Republic of Korea) implemented a system (called Safe Korea) supported by the Ministry of the Interior and Safety that collects a variety of per-sonal data in a centralized database in order to enforce quarantine orders and track possible contacts (M S Kim 2020) Israel also implemented a centralized involuntary data collection system for tracking COVID-19 cases and alerting those who may have been exposed (Hendrix and Eg-lash 2020) In Poland health authorities have set up mandatory ldquocheck-insrdquo involving a GPS-waypoint capture and ldquoselfierdquo photographs sent to the monitoring agency to ensure that individuals are not breaking quar-antine (Hamilton 2020)
These centralized systems can be designed to incorporate data from a variety of sources The data collected include location data from mobile phones QR codes can also be scanned to track the use of public transit where GPS data may be inadequate (due to low resolution) to accurately distinguish the occupants of one vehicle from another The data collected from mobile phones can then be integrated with data from other sources such as facial-recognition cameras credit card transactions and social media
38 Digital Contact Tracing for Pandemic Response
At the other extreme of technology invasiveness for contact trac-ing isolation and quarantine many corporations and working groups (including the AppleGoogle collaboration) have developed privacy-pre-serving proximity tracking (PPPT) using Bluetooth Low Energy (BLE) ldquohandshakesrdquo that record close contact between mobile phone users In most architectures these proximity data are stored in the usersrsquo phones as anonymized ldquobeaconsrdquo that cannot be used to re-identify the users directly If a user with a PPPT app installed on their phone tests positive and enters test results into their app those who have been identified as having been in close proximity to them can be notified by the app This notification can be automatic or at the discretion of the person who is COV+ depending on the app design If notified a user who has been in contact with a COV+ individual would receive a push notification alert-ing them to possible exposure (which may be timestamped) but with no other identifying information
Because of its reliance on anonymized data PPPT on its own is dis-tinct from manual contact tracing In recognition of this fact some de-signers and researchers now use the more descriptive term ldquoexposure no-tificationrdquo Moreover the public health usefulness of PPPT is uncertain it is unclear how PPPT can best be used in tandem with manual contact tracing especially if the data it collects are inaccessible to or unusable by public health authorities It remains to be seen whether PPPT will provide significant benefit operating alongside but not integrated into manual contact tracing
Between these extremes there are a number of possible middle-ground approaches that aim to strike a balance among public health utility tech-nological feasibility and user privacy protections This middle ground divides into two rough categories centralized storage of de-identified data and decentralized storage of personally identifying data The United Kingdomrsquos NHSX is reportedly developing an app that would utilize BLE handshakes to collect anonymized proximity data which would then be stored on a centralized government-operated server
The most prevalent middle-ground approach in the United States context involves the collection and storage of personal datamdashincluding identifying information and location datamdashon the userrsquos phone This de-centralized but personally identifiable data can then be voluntarily shared with public health officials if the user tests positive for SARS-CoV-2 For
Digital Technology and Contact Tracing 39
example an MIT team has developed an app called Private Kit Safe Paths (MIT nd) that stores usersrsquo location data on their phone for 28 days If a user tests positive she can voluntarily upload her location data to a website that is accessible only to public health officials Officials can then analyze this personally identifiable data and subsequently broadcast re-dacted and de-identified data to other users Healthy users would have access to these redacted location data of COV+ users but their own data would not leave their phones (The developers plan to incorporate BLE proximity data once available) Along similar lines the North Dakota state government has rolled out an app that stores both location data and proximity data on a userrsquos phone which can be voluntarily released by the user to public health authorities if the user tests positive (NDDoH 2020) At a minimum the storage of user location data can function as a ldquomemory aidrdquo if the user tests positive But releasing the data to public health authorities may help them analyze the spread of COVID-19 and alert individuals or groups that have been in contact with persons who are COV+ An overview of various DCTT apps and platforms as well as features that are relevant to this analysis are provided in Table 1
Because DCTTs are so new very little is known about their actual utility to public health authorities for controlling this pandemic Although multiple countries that have had success in greatly reducing transmission of SARS-CoV-2 have included DCTT in their response these countries have employed multiple simultaneous approaches to controlling the vi-rus including manual contact tracing and it is difficult to disentangle what made those responses successful Preliminary impressions from Ice-land may suggest that DCTT at least in that context had a small impact on reducing transmission ldquoespecially compared with methods of manual contact tracing such as phone callsrdquo (Hadavas 2020) This is with the highest public download rate of any DCTT app thus far
DCTTs have the potential to be helpful but they also have the poten-tial to distract from other public health efforts including manual contact tracing Concerns about implementation of DCTT from the public health perspective include that data generated may not be useful to public health authorities either because they donrsquot include detailed data to aid contact tracers or because the data are inaccurate (Mills Rodrigo 2020) DCTT if not calibrated well could be overly inclusive and create many false positives This would be harmful to those individuals being notified and
Purp
ose
Tech
nolo
gies
Use
dD
ata
Stor
age
Part
icip
atio
n
Proximity-based exposure notification
Digital contact tracing (DCT)
Bluetooth LE
GPS
SMS
Centralized
Decentralized
Mandator (actually or functionally)
Voluntaryopt-in
Gov
ernm
ent a
cces
sA
pp N
ame
Dev
elop
er o
r Cou
ntry
Max
WeC
hat
Alip
ayCh
ina
D
ata
com
es fr
om g
over
nmen
t so
urce
s lo
catio
n da
ta s
ent t
o po
lice
Intervention Type
Middle Ground
Trac
e To
geth
erSi
ngap
ore
M
anda
tory
gov
ernm
ent
acce
ss if
pos
itive
NH
SXO
xfor
dO
xfor
d
Gov
ernm
ent m
aint
ains
dat
a
Nex
tTra
ceFr
ed H
utch
inso
n Ca
ncer
Re
sear
ch C
ente
r
Gov
ernm
ent m
aint
ains
dat
a
but n
o st
orag
e
COVI
D
Safe
Path
sM
IT
Volu
ntar
y up
load
by
user
s
who
test
pos
itive
Aar
ogya
Set
uIn
dia
A
nony
miz
ed a
ggre
gate
Care
19N
orth
Dak
ota
In
agg
rega
te o
ptio
nal i
f pos
itive
Minimal
Covi
dSaf
eU
niv
of W
ashi
ngto
n
Non
e
Covi
dWat
chU
niv
of S
tanf
ord
amp U
niv
of W
ater
loo
To
val
idat
e te
st re
sults
CoEp
iCo
Epi
O
pt-in
to s
hare
BT
and
sym
ptom
lo
g w
ith C
oEpi
ser
ver
itoG
erm
any
N
one
pos
itive
resu
lts to
ito
serv
er
TAB
LE 1
Ex
ampl
es o
f Dig
ital C
onta
ct T
raci
ng T
echn
olog
ies
to S
uppo
rt A
ctiv
e Pu
blic
Hea
lth S
urve
illan
ce a
nd R
elev
ant F
eatu
res
Digital Technology and Contact Tracing 41
asked to quarantine unnecessarily and it could result in large proportions of the population remaining at home at any one time Individuals living or working in congregate settings could receive frequent notifications that would result in their inability to leave quarantine for long periods of time Finally public health authorities could also become inundated by data from these technologies and not have sufficient approaches to manage or analyze the incoming information
Relevant Differences between Manual and Digital Contact Tracing
There are several noteworthy differences between manual contact tracing efforts and use of DCTT First there is a significant amount of evidence regarding the effectiveness of manual contact tracing which is lacking for DCTT Second manual contact tracers interact with individuals who are confirmed or suspected cases and contacts of cases but not other members of the general public DCTT intervention would affect all users regardless of circumstances (though some more than others) Third manual contact tracing occurs most often through human-to-human encounters with the opportunity to clarify misconceptions address worries and express sympathy and other important affects DCTT can certainly incorporate sharing of important information and potentially communicate some af-fect but it currently lacks a range of other human capabilities and char-acteristics Fourth there typically are fewer data intermediaries in manual contact tracing (fewer entities handling data) in DCTT a valid argument could be made that a wide range of technology developers (and perhaps mobile network operators) must remain connected to relevant data in order to continuously identify problems and improve functionality
It is because of these and other differences that DCTT has been pro-posed as a potential complement to rather than a replacement for man-ual contact tracing However over time it is possible that technology could develop to close gaps between some of these differences (if and as needed) and in parallel the goals of contact tracing and public health surveillance may evolve
Ethics of Designing and Using DCTT
43
THREE
43
Those developing DCTT and those considering its use should systemat-ically take into account and document alignment with the guiding princi-ples outlined in this report
When considering the ethics of DCTT key ethical questions con-cern the features that DCTT should have (eg should digital contact tracing apps collect usersrsquo location data) whether and how individualsrsquo data should be shared with public health authorities how ethically to encourage use of DCTT (eg under what circumstances would it be eth-ical to incentivize or mandate use of DCTT) what kind of supports and equity-promoting measures should accompany use of DCTT and how governance and oversight of DCTT should be structured
The sections that follow consider these questions one by one A key conclusion of this report is that these features of the design and use of DCTT are ethically interrelatedmdashreaching a determination regarding any one question requires careful consideration of them all Rather than reaching ldquoone size fits allrdquo conclusions about specific features of uses of DCTT decision makers should ethically assess DCTT systems holistically
Generally a public health measure is ethically justifiable if it strikes a reasonable balance between competing considerations and if it pro-vides sufficient public health benefit (or the prospect of benefit) to justify the burdens associated with it DCTT systems are ethically justifiable if they strike a reasonable balance between multiple ethical considerations including
bull enabling an effective and efficient public health response
bull protecting individual privacy and preventing harms to individuals
44 Digital Contact Tracing for Pandemic Response
including harms from sensitive data being revealed and from erro-neously being subjected to isolation or quarantine orders
bull allowing individuals to control what information about them is collected and revealed to whom including through appropriate dis-closure and authorization processes for data collection
bull promoting equitable distribution of benefits and burdens of DCTT
bull maintaining public trust in DCTT and in the COVID-19 public health response and
bull taking seriously the future implications of decisions that we make today
To illustrate a holistic assessment consider whether it is ethically jus-tifiable for an employer to mandate that employees use a DCTT as a condition of returning to work This will depend upon many features of the DCTT system what kind of data the DCTT collects (eg does it collect location data or just record proximity events) whether there is public health capacity to make good use of these data what the data are used for (eg will the employer ban an employee from the workplace on the basis of a DCTT-identified contact) what kind of social supports are available (eg is there paid leave for employees) what employeesrsquo attitudes are toward use of DCTT and whether mandating use is likely to have public health benefit among other factors These factors may vary from place to place and may change over the course of the pandemic Thus there is no ldquoone size fits allrdquo ethically optimal approach to DCTT
Justifying the Use of DCTT Systems
A foundational issue is why deploying any DCTT during a pandemic is justified given there are manual contact tracing capabilities that are well established while the performance and effectiveness of novel technolo-gies is less established The need to move quickly to minimize the spread of the virus poses challenges here as the data needed to fully make the case that these technologies substantially contribute to the public health response may not be available prior to widespread use The primary ar-gument for DCTT is that the capacity of manual contact tracing may be
Ethics of Designing and Using DCTT 45
exceeded and we may not be able to bolster the public health workforce rapidly and sufficiently enough to meet needs DCTT has the potential to quickly and exponentially expand the reach of contact tracing In ad-dition DCTT may allow more efficient identification and quarantine of potential contacts of COV+ people than manual contact tracing alone particularly given the high number of infections that have been spread by asymptomatic individuals
Nonetheless reasonable people disagree about the prudence of pur-suing DCTT especially given its limited performance history and poten-tial risks including diverting attention and resources from more effective interventions The limited attention and resources available during a pan-demic must be allocated efficiently and effectively
To justify potentially widespread use of technologies such as DCTT therefore a number of considerations must be addressed
bull whether the technology is designed to meet an important and unmet public health need
bull whether there is sufficient evidence or reason to suggest that the technology will be effective at serving its purpose
bull whether the outbreak is characterized by sufficiently severe morbid-ity and mortality and a high rate of disease transmission to warrant large-scale introduction of novel systems
bull whether there are other less autonomy-restricting or less risky al-ternatives to widespread use and
bull whether it is reasonably likely that a sufficient number of individ-uals will use the technology to achieve the intended public health benefit
Monitoring and Evaluating Technologies to Inform Policy and Practice
A number of public health ethics principles necessitate the ongoing mon-itoring and evaluation of DCTT systems First DCTT must be shown to perform reasonably well at achieving its stated goal reducing the spread of SARS-CoV-2 The effectiveness of DCTT programs should be illus-trated at a number of stages
46 Digital Contact Tracing for Pandemic Response
1 Robust initial technology testing is needed to publicly justify the widespread adoption of DCTT and avoid public failures which may hamper future uptake (eg Lovejoy 2020 Morse 2020) This typically includes alpha testing in virtual environments and beta testing in different community settings
2 If and when a DCTT is implemented on a wide scale it must be monitored on an ongoing basis to assess reach effectiveness func-tionality best practices and any harms
3 When approaching a previously identified stopping point for use of DCTT monitoring can help to identify when utilization is no longer needed
If at any of these points evidence clearly suggests harm (particularly in comparison to other methods that the public might find more acceptable) this evidence should provide a basis upon which to revisit strategies pri-orities and allocation of resources Attention should be given to foresee-able side effects that may dramatically influence the overall effectiveness of the program such as individuals carrying their smartphones around with them selectively so as to avoid particular undesired consequences of DCTT policies
Anonymized aggregate data including user feedback must be eval-uated to ensure that benefits and burdens are distributed fairly As noted earlier unintended burdens may include inequitable outcomes that may arise in a DCTT program for example resulting from uneven access to the required technology to participate disparate concerns about sur-veillance within some communities that might limit widespread use or discrimination that may result from being identified as COV+ due to the program or for communities that are termed ldquohotspotsrdquo based on maps of COV+ location data Additionally it is possible that some communi-ties might get higher rates of false positives because they are located in densely populated areas thus increasing the burden of self-isolation If any of these inequities are identified steps must be taken to mitigate them
Finally numerous actors should engage in the monitoring and eval-uation of DCTT systems Technology developers and public health re-searchers have a clear role in this process Technology developers should work with public health researchers to monitor accuracy precision func-
Ethics of Designing and Using DCTT 47
tionality confidence of estimates sources of error and the like Research-ers may also be able to contribute innovative methods to systematically and rapidly evaluate candidate technologies such as by deploying cluster randomized stepped wedge (Hemming et al 2015) or adaptive trial de-signs and techniques (eg response-adjusted randomization) (Pallmann et al 2018) These approaches were also proposed for use in research to assign candidate experimental treatments and vaccines during the 2014-15 Ebola outbreak (Berry et al 2016) When formal research activities are pursued ethics principles and legal requirements for the conduct of research should apply (eg The Belmont Report)
Furthermore any workplace or institution that incentivizes or man-dates use of DCTT has a responsibility to provide evidence that the in-tervention at minimum is not likely to cause harm and to monitor for unanticipated burdens In all cases it is vital that a trusted intermediary be involved in the evaluation of DCTT programs to limit perceptions of bias and ensure a legitimate basis for decision-making Nonsensitive aggregate DCTT analyses should be made available to the public so as to permit verification and inform continuing public debates about its useful-ness and necessity At an individual level data should be available to users that would permit them to further investigate their personal risk with public health officials or other health workers This is important not only to ensure their health and well-being but also to add a layer of protection against unnecessary quarantine
Recommendations
bull Reviews of DCTT systems must be conducted in part by an inde-pendent intermediary that has established the publicrsquos trust
bull Those who authorize use of DCTT within a particular jurisdic-tion or institution should continuously and systematically monitor the technologyrsquos performance in that context This should include monitoring for effectiveness and benefit monitoring for harms and monitoring for the fair distribution of both benefits and harms They should also monitor evidence that is being generated in other contexts about their selected technological solution and about other competing technologies
48 Digital Contact Tracing for Pandemic Response
bull Data should be available to users that would permit them to further investigate their personal risk with public health officials or other health workers to add a layer of protection against unnecessary quarantine
Public Trust and Public Attitudes
Researchers have estimated perhaps conservatively that DCTT use by 80 of smartphone ownersmdash56 of the populationmdashwill be needed to suppress the epidemic (Hinch et al 2020) These estimates also highlight that some decrease in transmission would be realizable even with lower rates of technology adoption As such in order to maximize impact it is essential to gain a thorough understanding of public perspectives on DCTT including which features and uses of the technology the pub-lic finds acceptable which kinds of DCTT the public would be most likely to use and which designs and uses of DCTT would maintain or jeopardize public confidence and trust There will be variation in public attitudes within and across societies and over time
With respect to what we currently know about public attitudes and trust in DCTT in the United States polling data suggest some potential support and also some divisions regarding willingness to use the technol-ogy Polls conducted by groups based at the University of Zurich (Hargit-tai et al 2020) and the University of Oxford (Altmann et al 2020) suggest that more than 60 of Americans would be willing to install such an app Both a Washington PostndashUniversity of Maryland poll (2020) and a Kaiser Family Foundation poll (Kirzinger et al 2020) show roughly half of the population would be willing to install the app Over half of the population (59) would be willing to share their COVID-19 positive test result with an app in order to anonymously share that information with their contacts (Washington PostndashUMD 2020) Only 29 of respondents to a March 12ndash27 Oliver Wyman Forum poll (Elliott et al 2020) said that they would be willing to share their location data Additionally Washing-ton PostndashUMD data and Pew data from 2019 suggest that approximately one in six Americans do not have a smartphone and thus cannot use the technology without intervention (Pew Research Center 2020)
People may be more willing however to download an app if it will
Ethics of Designing and Using DCTT 49
ease social distancing policies and allow for more economic and social activity Willingness to install a contact tracing app increased among re-spondents to the Kaiser Family Foundation poll from 50 to 66 when respondents were asked if they would be willing to do so to allow schools and businesses to reopen Additionally who develops or administers the app appears to matter Respondents to the Washington PostndashUMD poll indicated higher levels of trust that their anonymity would be preserved by public health agencies and universities than by tech companies or health insurance companies Further more respondents to the Oliver Wyman Forum poll were willing to share their health information with public health authorities (55) than the local government (35) their employer or school (33) or the federal government (27)
These data suggest that people will be more willing to use a contact tracing app when the potential benefits are clearly identified and valued such as lifting social distancing measures and they will be more willing to do so if the data are going to a public health agency rather than the federal government or a tech company Other factors that seem to be as-sociated with greater willingness to install a contact tracing app include younger age and the app source (Hargittai et al 2020) with a preference for apps distributed by public health agencies over others such as health insurers or public universities (Hargittai and Redmiles 2020) However all of this must be read with caution as public polling may not be repre-sentative of some populations or of widespread public attitudes Further these attitudes may shift over time and may be discordant with behaviors (Barth and de Jong 2017)
Deliberative public engagement efforts would be an appropriate means of filling in gaps in understanding about the acceptability of dif-ferent approaches (Fishkin and Laslett 2003 Cavalier 2011) In addi-tion including the public particularly in the earlier stages of planning a path to sustainable resolution to the pandemic could serve to help disseminate a nuanced understanding of what is at stake including the key challenges and trade-offs Aggregated public polling results are not sufficient as a proxy for careful analyses of the ethical challenges but they do provide a necessary input for these analyses Integrating lessons and outputs from public engagement into guidance and other products requires special attention and should be validated and enhanced through further engagement
50 Digital Contact Tracing for Pandemic Response
Recommendations
bull More research into public attitudes is needed In particular in-depth qualitative research should examine public attitudes about percep-tions of trust in DCTT among different communities which fea-tures of DCTT influence trust and the extent to which people are willing to provide different types of data through DCTT to help their community
bull States and localities that are considering adopting DCTT should engage with the public to increase their understanding of the ac-ceptability of DCTT design features and uses among diverse communities
Designing Flexible Technology to Maximize Public Health Utility While Respecting Other Values
Values in Design
Efforts to advance DCTT in the United States and elsewhere have empha-sized the importance of ldquoprivacy by designrdquo that is building privacy and security protections into the design of technology rather than counting on responsible use alone (Cavoukian 2010) As noted above some major technology companies have signaled this position through development of decentralized privacy-preserving proximity tracking (PPPT) systems These systems embed features such as decentralization anonymity of us-ers bans on collection of location data and minimal reliance on or inte-gration of public health authorities or other government actors Many of these features have also been embraced early by advocacy organizations (Crocker Opsahl and Cyphers 2020 Electronic Privacy Information Center 2020 Kahn Gilmor 2020) and in an open letter from nearly 300 researchers (ldquoJoint Statement on Contact Tracingrdquo 2020)
Privacy by design provides principles that incorporate one set of val-ues (privacy) into the design of DCTT Importantly the principles ac-knowledge the need to design privacy defaults into systems while main-taining the capacity of those systems to achieve their otherwise justifiable ends Put another way privacy by design ldquoembraces legitimate non-pri-vacy objectives and accommodates them in an innovative positive-sum mannerrdquo (Cavoukian 2010 p 4)
Ethics of Designing and Using DCTT 51
This stance simple in its statement is not easy to satisfy Given that ldquoobjectivesrdquo are themselves driven by values it begs for an articulation of additional values (aside from privacy) that individuals and groups within societymdashincluding many privacy advocatesmdashmay believe to be important For example at any moment in addition to valuing their own privacy individuals may value efficiency equity autonomy economic well-being companionship patriotism or solidarity Moreover the above stance necessitates an acknowledgment that peoplesrsquo value priorities often change when circumstances change not least of which during a pandemic when mass physical distancing has made it difficult to fully realize many important values (aside from physical privacy) A different orientation is needed at this moment As Flanagan Howe and Nissenbaum (2008) conceptualized in 2008 we should take a ldquovalues in designrdquo approach to DCTTmdashan approach that designs a broader range of values such as those enumerated above into technology
This approach requires a wider ethical lens through which to ex-amine DCTT and requires hard but important work to appropriately balance competing interests within technology architecture For example there is value in technology providing users the option to collect their location history and share it with public health professionals in order to advance the public health response increase system efficiencies (eg by contributing information that can lead to better data processing) and reduce the burden on essential workers For some this might be an ex-pression of autonomy solidarity or patriotism At the same time there is value in further advancing autonomy by designing technology to allow individuals to control what data about them are collected and shared
Justifying a Middle-Ground Approach to DCTT
We ought to embrace a DCTT that has a default of interoperability and privacy protection but that does not stop there Triggering events such as entry of a positive test result or receipt of a notification that one was proximate to someone who tested positive could for example generate a push notification that users can acknowledge in order to permit transmis-sion of potentially useful location data to public health authorities This could be accompanied by an explanation of the value of the information and relevant restrictions on its use
At this point it is worth reiterating that manual contact tracingmdash
52 Digital Contact Tracing for Pandemic Response
which involves collecting information from people whorsquove tested posi-tive and their contactsmdashincludes collection of personal information and potentially embarrassing or sensitive data about the places theyrsquove been and the people theyrsquove had contact with Manual contact tracing efforts use these data to uncover ongoing transmission provide useful informa-tion tailored to the individual and enable isolation and quarantine as necessary
It stands to reason that if these forms of data can be collected by a DCTT and provided to public health authorities in a maximally secure and voluntary way (with clear rules regarding authorized uses) this may amplify public health authoritiesrsquo manual contact tracing efforts For ex-ample location data from DCTT could help jog peoplersquos memories about where theyrsquove been and fill in memory gaps This is especially relevant given the long period of infectivity of SARS-CoV-2 which begins before people are symptomatic and therefore before they are aware they are infected (Ferretti et al 2020) Location data might reveal that a COV+ person was at a restaurant at an exact time and date which could be followed up by contact tracers who could alert the public or use other measures to reach those who were also present in the restaurant at the same time In other disease contexts (see Furlanello et al 2002 Dredze et al 2013 Eckhoff and Tatem 2015 Fraccaro et al 2019) geolocation data have demonstrated some potential to support epidemiology and dis-ease surveillance with technical cautions regarding accuracy and the like (Beukenhorst et al 2017)
These benefits are currently speculative for DCTT At present pro-viding public health authorities with large amounts of data on cases and potential case contacts will be useful only if there is sufficient public health capacity to follow up on these data In addition there is a risk of low-quality data from DCTT flooding the system Investigating poten-tial case contacts identified by a DCTT may distract them from other important efforts and at some point overwhelm public health capacity altogether Whether and to what extent data from DCTT will benefit contact tracing efforts is unknown pointing again to the importance of continuously collecting high-quality evidence about DCTT
Nevertheless what would enable the most flexible and potentially robust public health response is to design DCTT so that restricted data sharing is possible From an ethics perspective the collection and use
Ethics of Designing and Using DCTT 53
of sensitive data in manual contact tracing efforts (described above) is typically seen as ethically justifiable so long as there is sufficient public health benefit and need Thus wouldnrsquot it seem appropriate from both a public health and ethics perspective to design DCTT systems to enable similar data to be shared with public health authorities when and if there is ethical justification for sharing them
Why instead do so many advocate that DCTT should be designed as a ldquominimalrdquo system when this arguably ties the hands of public health and individual users and precludes the collection of data that public health authorities (and indeed many other apps on our phones) typically collect We here consider and appraise some of the reasons that may motivate individuals and groups to argue for minimalistic positions
1 Proponents of minimal systems may believe that such systems will be most
widely adopted Some groups have maintained that only these systems will earn and maintain public trust and be widely adopted (Simpson and Conner 2020) For example the previously referenced open letter (ldquoJoint Statement on Contact Tracingrdquo 2020) asserts ldquoSome of the Bluetooth-based proposals respect the individualrsquos right to privacy whilst others would enable (via mission creep) a form of government or private sector surveillance that would catastrophically hamper trust in and acceptance of such an application by society at large It is crucial that citizens trust the applications in order to produce sufficient uptake to make a difference in tackling the crisis It is vital that in coming out of the current crisis we do not create a tool that enables large scale data collection on the population either now or at a later time Thus solutions which allow reconstructing invasive information about the population should be rejected without further discussionrdquo
Response While it is true that public trust in and acceptance of DCTT is essential for its success there is insufficient evidence that public trust would be threatened by a DCTT system that has the capacity to collect location data and enable voluntary sharing of those data with public health authorities A contrasting perspec-tive is that maintaining public trust requires maintaining public confidence that the DCTT system is providing useful information is benefiting and not harming individuals and is advancing the
54 Digital Contact Tracing for Pandemic Response
public health response (Leprince-Ringuet 2020) From this per-spective a system that is less well integrated into the broader public health response or that generates a higher rate of false positives (as some suggest decentralized approaches might (Fraser et al 2020)) may fare worse when it comes to maintaining public confidence and trust
2 Proponents may hold the view that minimal systems are harmless (or nearly
harmless) to individuals This is because individuals are anonymous none of their location data are gathered and none of their identifiable data are shared with anyone In contrast DCTT systems that collect and share identifiable data including location data may be seen as posing risks of harm to individuals
Response While minimal systems may be harmless (or nearly harm-less) from the perspective of protecting privacy they may not be harmless from the perspective of public health if they generate system inefficiencies through producing too many false positive or false negative contacts Aside from presenting a challenge for public health professionals false positives could also harm individ-uals If users receive a large volume of automated messages alert-ing them to proximity events will this cause distress Will a large volume of alerts cause users to become disengaged and stop using the DCTT or lose confidence in contact tracing more generally as a legitimate method of disease control Admittedly these are just potential harms and risks it is unknown the degree to which they will materialize The point is that privacy-related harms are not the only relevant harms to individuals that we should consider when assessing DCTT
We acknowledge the risk under a middle-ground DCTT of data being used in ethically unjustifiable and harmful ways For exam-ple it would be against the principles and recommendations artic-ulated in this report for data to be sold or monetized by technology companies or others for corporate gain and this misuse of data would be more intrusive if the data were potentially identifiable What makes it ethically justifiable to take this risk is the compen-sating benefit of allowing the most flexible and robust public health
Ethics of Designing and Using DCTT 55
response during the pandemic but this alone is not sufficient The risk of inappropriate uses must be reduced by ensuring stringent requirements for data security and access as well as clear legal protections and recourse for any violations (as discussed further below)
3 Proponents may believe that DCTT systems should not collect location data
as this would be too intrusive and of insufficient value Some proponents of PPPT systems maintain that recording proximity events is sufficient and data relating to usersrsquo movement and location should not be collected (Ingram 2020) The thought may be all we need to know is whether two individuals came into close enough contact for viral transmission to have occurred we donrsquot need to know where or when this contact occurred and there is no need to collect and store usersrsquo location data
Response This conclusion might be too hasty As discussed above there is potential (though unproven) benefit to providing public health authorities with location data Location data could help jog peoplersquos memories about where theyrsquove been provide more context for understanding the nature of ldquoproximity eventsrdquo captured by the DCTT and allow public health authorities to quickly define a category of individuals who may be at risk Collecting location data from cases is what public health authorities do on a regular basis following best practices for manual contact tracing
In addition many peoplersquos location data are currently gathered by apps on their phones and used for various purposes such as to provide more accurate navigation to offer entertainment or to improve services Many are willing to accept these capabilities because they provide some value in return Why not allow DCTT to also collect these data so that the data are available for users to share with public health officials who can then do their work more effectively and refine their understanding of how the disease transmits If many are willing to have these data used to find a bet-ter route home why not let individuals share these data to support the effort to save lives
56 Digital Contact Tracing for Pandemic Response
4 Proponents may hold the view that minimal systems pose little or no threat
to individual autonomy whereas systems that collect identifiable data and
integrate public health do pose a threat to individual autonomy For exam-ple they may worry that use of DCTT could be mandated and not a voluntary choice and in this circumstance mandatory use of minimal DCTT would be less intrusive risky and privacy violating Another worry might be that itrsquos theoretically possible that DCTT could share individualsrsquo data with public health authorities without usersrsquo full understanding if the technology does not even gather identifiable data then itrsquos not possible for these data to be shared without the individualrsquos consent
Response We discuss the importance of appropriately designed disclosures and consent below as well as the high bar that would need to be met to ethically justify mandatory use At this time mandated use of DCTT by states or institutions is not justifiable given uncertainty about potential harms and benefits Users should have a meaningful opportunity to review and understand infor-mation about the specific technology and its uses and to consent Assuming that individuals are not required to use DCTT and that they provide consent to using it designing DCTT to make data collection and sharing possible is the design choice that maximizes individual autonomy because it provides individuals with options they may value
Individuals may wish to share their data with public health au-thorities for both self-interested and altruistic reasons For exam-ple someone who has tested positive for SARS-CoV-2 and enters this test result into an app may wish to be connected to public health authorities in order to be provided with needed information resources and support She may wish for public health authorities to be provided with her phone number in case they need to reach her to provide additional information Further someone who has been alerted by an app that he had a ldquoproximity eventrdquo with a person who has tested positive for SARS-CoV-2 may wish he had location data to share with public health authorities in order to help ascertain whether this event is a cause for concern or whether it is likely a false positive (eg he and the COV+ person were sepa-
Ethics of Designing and Using DCTT 57
rated by a wall) Someone who tests positive for the virus may also wish to share their location history with public health authorities in order to be as helpful as possible to the overall public health re-sponse by facilitating de-identified aggregate analyses that identify locations of higher transmission or contribute to refining overall understanding of the disease and pandemic
5 Concerns about ldquosurveillance creeprdquo and the long-term downstream effects of
digital contact tracing system may also motivate embrace of minimal DCTT Digital contact tracing technology that collects identifiers and loca-tion data and has the capacity to share them with public health au-thorities may represent a massive and concerning increase in govern-ment surveillance of the public It might be feared that the use of this surveillance capacity in the COVID-19 response sets an unwelcome precedent for future use in other contexts Designing DCTT as min-imal systems may be a way to minimize the risk of surveillance creep and to minimize the harms associated with potential future uses of the technology
Response Surveillance creep is a serious concern To guard against surveillance creep protections should be put in place to ensure that only those data that are necessary and relevant for the public health purposes at hand are collected and used and data should be kept only for the period of time needed for those public health purposes In the face of these concerns it is important to emphasize that widespread use of DCTT in the COVID-19 response is justi-fied by the exceptional circumstances of the current pandemic and their use in this context does not imply that future public health use is ethically appropriate without significant public debate (eg use in seasonal flu surveillance efforts) Future use will require in-dependent justification Use of DCTT in other contexts (eg law enforcement or immigration enforcement) is also presumptively unethical
All in all the arguments that DCTT should be designed as a minimal system are not convincing Rather DCTT should be developed through a ldquovalues in designrdquo approach with a core set of features that protect pri-
58 Digital Contact Tracing for Pandemic Response
vacy with enough flexibility to be used differently depending upon local conditions evolving evidence and individual preferences What kind of digital contact tracing system will strike the right balance between public health goals and other considerations will depend upon circumstances For example whether it is even beneficial to provide public health au-thorities with volumes of data about potential contacts of COV+ people will depend in part upon whether they have the capacity to make good use of those data This will vary from location to location and will change over time
Recommendations
bull Technology companies should not alone control the terms condi-tions or capabilities of DCTT nor should they presume to know what may be acceptable to members of the public
bull A ldquovalues in designrdquo approach to development of DCTT should be adopted (Flanagan Howe and Nissenbaum 2008 Knobel and Bowker 2011) Robust public- and user-engagement activities should be pursued to identify and incorporate to the extent pos-sible a range of values into the design of the technology These values may include privacy but also autonomy efficiency equity or others Technology design should reflect an appropriate balance and prioritization of identified values
bull Technology design should not be static but rather it should be ca-pable of evolving depending upon local conditions new evidence and changing preferences and priorities
bull DCTT should be designed to have a base set of features that protect privacy with layers of additional capabilities that users may choose to activate An initial default should be that user location data are not shared but users should be provided with easy mechanisms and prompts to allow for opting-in to this capability with encourage-ment to the public if and as it is shown to be critical to achieving public health goals
Ethics of Designing and Using DCTT 59
Policy Positions to Advance Widespread Use of Digital Contact Tracing Technologies
The public health value of a DCTT depends in part on the number of people who use it This section concerns broad public policy positions that relate to the widespread adoption of DCTT What are ethical means of encouraging or securing widespread adoption of DCTT systems Un-der what circumstances would it be ethical to mandate their use or incen-tivize their use What enforcement challenges exist
Mandating Use
Digital contact tracing has occurred without the publicrsquos explicit volun-tary agreement in some countries such as China and Israel In others use has been voluntary (Valentino-DeVries Singer and Krolik 2020) For example Singapore adopted an app that the public could use on a vol-untary basis and approximately 20 of the population has downloaded and used it Norway has recently launched a contact tracing app that was downloaded by roughly 30 of the population in the first week that it was made available In the United States many advocates and researchers have argued that use of digital contact tracing tools must be fully volun-tary this is the dominant perspective
There are numerous ways that DCTT could be put into use without user choice For example as has been done in Israel location data from mobile phones could be collected and used by the government without usersrsquo consent Use of an app could be formally mandated as a precon-dition for returning to work or school or even further to control entry into a facility or onto transportation such as airplanes through scanning of a QR code to demonstrate personal exposure levels (Gan and Culver 2020)
While these approaches are hard to imagine in the United States some contend that mandatory use of digital contact tracing tools could be ethical and may even be ethically required Mandating use of digital contact tracing tools could in theory vastly increase the effectiveness of digital contact tracing systems and thus may save more lives and allow states to lift lockdowns sooner or avoid reimposing lockdowns in the future Canca (2020) argues that use of privacy-by-design digital contact
60 Digital Contact Tracing for Pandemic Response
tracing tools should be mandatory because the use of these tools will be nearly harmless if there are sufficient privacy protections In addition mandatory use of DCTT that embraces these principles is significantly less intrusive at the individual level than manual contact tracing which involves the collection of personally identifying and potentially sensitive data In this light it could be argued that such mandates are actually pref-erable from the perspective of both public health and individual liberty insofar as they reduce the likelihood of ldquostay at homerdquo orders which are a severe limitation of individual liberty
Nevertheless mandated use of DCTT systems faces considerable ob-stacles For example people may not adhere to the mandate by simply leaving their phone at home thus preventing their activities from being tracked Even more harmful would be if people react to a mandate and a perceived violation of liberty and privacy by employing location and Bluetooth spoofing software to shield their real contacts behind a screen of misinformation The introduction of this misinformation into a contact tracing effort might severely undermine its effectiveness The possibility of nonadherence also raises the issue of enforcement would high rates of nonadherence be permitted or would enforcement be attempted (if even possible) Perhaps more important should the technology not deliver the hoped-for benefits having mandated the use of an unproven technology could result in a loss of public trust in the technology the entity instituting the mandate and potentially the larger public health response (Bernstein et al 2019)
Mandatory DCTT could also be used to enforce quarantine restric-tions and stay-at-home orders for those who are COV+ or are determined to be at heightened risk The use of DCTT in enforcement activities raises a number of ethical (and legal) issues that are beyond the scope of the present analysis In particular individuals have a heightened interest in personal privacy if their data can be used to restrict their freedom of movement and other civil liberties At a minimum stringent procedural protections would be required to ensure that the data collection is fair and unbiased and that DCTT users are provided with adequate informa-tion in advance about how their data may be used
Mandatory use policies for DCTT must therefore convincingly ad-dress a number of questions including
Ethics of Designing and Using DCTT 61
bull Is the technology designed to meet an important and unmet public health need
bull Is there sufficient evidence to suggest that the technology will be effective at serving its purpose
bull Is the outbreak characterized by sufficiently severe morbidity and mortality and a high rate of disease transmission
bull Are there other less autonomy-restricting or less risky alternatives to widespread mandatory use of DCTT
bull Is it possible and likely that a sufficient number of individuals will comply with a mandate
bull Can inequities in the burdens and benefits of the mandate be suffi-ciently addressed through social protections and countermeasures
bull Can enforcement and enforcement discretion be implemented in a manner that is consonant with fundamental rights
bull Will those subject to the mandate interact closely with a population that is at high risk of morbidity or mortality if they contract the virus
bull Is it possible to mandate use and remain consistent with important ethical and legal principles
These questions would need to be satisfactorily addressed and explicitly documented by any decision maker considering mandatory use includ-ing government officials institutional leaders and employers Particularly important is the need to identify reliable evidence that the DCTT would be effective and to ensure that the burdens and benefits of use are equita-ble and justifiable If use of a DCTT is a condition for returning to work or school in person those who refuse or are unable to use DCTT should not lose their jobs or positions as a result and adequate support should be in place for people who are asked to self-quarantine
Finally it is important to distinguish a mandate from a ldquopushedrdquo program installation or a default setting in an application which can be modified by users A mandate relates to a policy of required use whereas the pushed programs or default settings relate to the chosen architecture for download and operation of the application
62 Digital Contact Tracing for Pandemic Response
Incentivizing Use
Perhaps the most effective way to generate widespread adoption of DCTT in the United States is to offer incentives to individuals who choose to adopt and who properly utilize the preferred DCTT approach in a voluntary system External incentives may help ldquonudgerdquo populations toward desired adoption targets Given the importance of widespread use of DCTT modest incentives ought to be considered for DCTT in the US if and when there is sufficient evidence of the technologyrsquos utility Note that in other contexts studies have shown that the provision of some incentive leads to an increase in adoption or utilization of public health programs (Singer and Ye 2013 Lee et al 2014) Moreover even a relatively small incentive can achieve much greater rates of adoption with some studies demonstrating that the incremental adoption gain de-creases as the incentive gets larger (Thornton 2008 Gibson et al 2019) In the context of COVID-19 incentives that might be both effective and ethically acceptable could include a relatively small monetary token free or discounted mobile phone service for a period of time or credit to be used by means of a mobile phone
Not all incentives are ethically appropriate For example making access to lifesaving health care contingent on using a DCTT or making valuable disease information available only to DCTT users would not be ethically appropriate In addition incentives cannot be used to over-come otherwise ethically unjustifiable technology design for example they should not be used as an offset for providing personally identifiable health information to other users
Importantly incentivization schemes must be kept distinct from man-dates as the latter require greater ethical justification To offer an incen-tive is to offer something of actual value to individual participants over and above what they are reasonably entitled to at baseline For example making a return to work contingent on using DCTT is not offering an incentive but instead imposing a mandate and it would have to be justi-fied as a mandate
In the context of COVID-19 it is also necessary to recognize that there is an inherent ldquoincentiverdquo behind the technologymdashthat is the prom-ise of more lives saved faster pandemic recovery and the reduction or elimination of blanket physical distancing Effective public communica-
Ethics of Designing and Using DCTT 63
tion of these goals if and when there is sufficient confidence in the tech-nology is important
Encouraging Use
Another important approach to increasing use of DCTT in the United States is for trusted leaders to encourage their use Community leaders public figures health care professionals and other respected individuals who have the publicrsquos trust and goodwill could be enlisted to commu-nicate with the public about DCTT and encourage its use drawing on notions such as communal responsibility solidarity and so on These en-couragements could be combined with other approaches (eg small in-centives) to optimize reach while continuing to respect individual choice
Recommendations
bull DCTT use should not be mandated at this time given uncertainty about potential harms and benefits Additional technology user and real-world testing is needed
bull Incentives can be a useful complement to encouragements how-ever any incentives for users to install and use DCTT must be equitable should not be coercive and should align with effective use of the technology (ie they should not incentivize downloading an app but then leaving onersquos phone at home)
bull Trusted leaders should be enlisted to communicate effectively with the public about DCTT and encourage its use should the technol-ogy demonstrate some potential The limits of knowledge regard-ing effectiveness should also be explained along with what will be done to improve technological capabilities as understanding evolves
Disclosure and AuthorizationConsent
In deciding whether to use DCTT voluntarily individuals must be suf-ficiently informed both through broad coordinated public engagement campaigns and individual-level disclosures and there must be a meaning-ful mechanism for users to consent It is important to recognize that while
64 Digital Contact Tracing for Pandemic Response
informed consentmdashwhich is characterized by detailed consent forms and requires a witnessed signaturemdashis the standard for most research and clinical care encounters (Faden and Beauchamp 1986) it is not typically the standard for public health disease surveillance In the public health context other relevant protections (such ethics training for public health professionals and strict data handling and confidentiality requirements) are in place and there is a strong public health interest in collecting the relevant data A more limited role for consent has been recommended for public health surveillance based on a reciprocal obligation of members of society to contribute to a ldquocommon goodrdquo and particularly in the con-text of a pandemic practical considerations such as time constraints and exigencies such as increasing morbidity and mortality (WHO 2017)
Under current circumstances given that (1) many individuals have time and capacity to consent (2) DCTT is being considered as part of plans for longer-term restabilization (3) DCTT is not a familiar part of our public lexicon (4) remote consent disclosure and authorization can be easily embedded in DCTT systems (Moore et al 2017) and (5) there are justifiable public deficits in trust with respect to various government and corporate actors handling potentially personal digital information a strong ethical case can be made for requiring a carefully crafted version of what is sometimes referred to as simple consent Simple consent consists of basic disclosure and voluntary agreement or authorization (Ali et al 2017) Three questions then arise
1 What information should be disclosed to potential users of DCTT
bull Information disclosed might include
deg Entity responsible for the technology
deg Its purpose
deg How it works (in lay terms)
Some participatory disease surveillance systems (eg Flu Near You) have received for-
mal ldquowaiversrdquo of consent requirements from institutional review boards (IRBs) in the US
As they undergo development these digital surveillance systems often straddle a line be-
tween public health surveillance and research hence the frequent need or desire to obtain
ethical review by an IRB (Ali et al 2019)
Ethics of Designing and Using DCTT 65
deg What users need to do
deg Any user options eg
Sharing geolocation data with public health authorities when that would facilitate a defined public health goal
Sharing de-identified metadata with technology develop-ers (for system enhancement)
deg User rights
deg How data will be handled
What data are collected
What data are shared (and how and with whom)
Purposes for which data can be used and not used
How data are secured and protected
Whether and what data will be retained (or will be deletable)
deg Potential benefits and any known risks
deg How to obtain answers to questions about the technology and public health response
2 How should this information be presented
Information should be presented leveraging eConsent models that are more accessible than long ldquoclickwraprdquo disclosures typical of mobile apps (Iwaya et al 2019) For example a simple open-source smartphone con-sent module that has been developed by Sage Bionetworks for research uses could be adapted to the public health surveillance context and to DCTT (Doerr Suver and Wilbanks 2016)
bull Formatting recommendations include (cf Doerr et al 2016)
deg simple and straightforward information
deg deliberately organized content
deg multimodal learning (eg visual audio written)
deg accessibility for disabled users
66 Digital Contact Tracing for Pandemic Response
deg multilingual text
deg engagement through interaction (eg swiping to navigate forward and backward)
bull The same simple information should be made publicly available via multiple other platforms (eg on websites in newspapers over social media)
bull More detailed disclosures should be made readily accessible to those who wish to learn more with no hidden surprises
3 How should users signal that they agree to the details specified in disclosures
Opt-in Models
Opt-in models are those that through an affirmative act such as clicking a button users would indicate their intention to use a DCTT This ap-proach is consistent with other app downloads where app details and privacy policies are made available through a download page and users are required to affirmatively click a button to install an app Once in-stalled some apps further alert users to particular ways in which phone capabilities or data will be used with some permitting selective toggling (opting-in or opting-out) of certain features With DCTT apps in addi-tion to disclosures provided on a download page the user could be guided through a simple interactive module embedded in the app (such as is described above) in order to increase the chance of meaningful exposure to important information about the technology and how data will be handled At that point any user options such as those itemized above could be described and choices made
Opt-out Models
There are at least two different ways in which the term ldquoopt-outrdquo has been used in this context The conventional use of the term ldquoopt-outrdquo is characterized by an act which signals an individualrsquos intention to decline something that would have otherwise occurred without intervention A few others have used the term to refer to ldquorevocation of consentrdquo for example the United States COVID-19 Consumer Data Protection Act of 2020 Senate bill (S3663) would establish a default opt-in positionmdash
Ethics of Designing and Using DCTT 67
requiring ldquoaffirmative express consentrdquo for collection and use of prox-imity and other related datamdashand refers to individuals having a right to later revoke their consent through an ldquoopt-outrdquo The latter use of the term is not our focus here
Given this a DCTT app that is voluntarily downloaded through an affirmative act would be difficult to characterize as an opt-out approach This leaves more passive surveillance systems that rely on automatic in-stallation of self-activating technology onto users phones There are a range of views among the authors of this report about the value of an opt-out approach for DCTT with some arguing for an opt-out approach on grounds that it might increase coverage and would be ethically acceptable if accompanied by similar disclosures as above to ensure users are aware of the technology and data uses (Mello and Wang 2020) This approach would present users with a mechanism to opt-out if they wish which should be reasonably easy to effectuate Under these circumstances as noted above an ldquoopt-outrdquo would not be synonymous with mandating use of the technology
Others among the authors argue that there is reason to believe that opt-in approaches may be able to sufficiently achieve desirable levels of utilization relative to opt-out approaches Unfortunately data related to opt-in versus opt-out models of DCTT are very limited One recent sur-vey (Altmann et al 2020) found that across five countries (UK Germany France Italy US) slightly more people reportedly would download an app under an opt-in system (748) than would keep an app on their phone under an opt-out system (677) Moreover when US respondents were directly asked which approach they would prefer 60 indicated a preference for opt-in This remained true across various demographic variablesmdashgender region political affiliation lockdown status and other characteristics Whether actual behaviors would align with anticipated behaviors in the context of DCTT remains an unanswered question that should be carefully studied under real-world conditions There are a range of important empirical questions regarding how much and what kind of impact (positive or negative) various types of defaults might generate for public health and for different mobile phone user groups including vul-nerable and marginalized users
Opt-out models for app authorization may encounter greater legal
68 Digital Contact Tracing for Pandemic Response
and political challenges especially if the COVID-19 Consumer Data Pro-tection Act of 2020 (S3663) the competing Public Health Emergency Pri-vacy Act (S3749) or another similar bill is enacted in the United States Both of these standing bills require affirmative opt-in consent Opt-out approaches also risk negative reactions from some mobile phone users a small number of whom may go so far as to intentionally interfere with data because of the perceived intrusiveness of an automatically installed tracking platform (Dixit 2020)
Given these considerations and the apparent willingness of a large portion of the population to opt-in to use DCTT an opt-in approach to authorization should be instituted to accompany initial DCTT rollout The feasibility and value of opt-out approaches should continue to be evaluated informed by what is technologically possible what local as-sessments of benefits and harms of the technology reveal over time and our evolving understanding of the degree to which an opt-out approach is likely to increase or decrease utilization Opt-out approaches should not be precluded
Recommendations
bull A clear and concise module consisting of basic disclosure and vol-untary authorization should be developed to accompany DCTT This module should not take the form of ldquoclickwraprdquo terms of service or end-user agreements but rather provide only essential information necessary for an individual to make a decision More detailed disclosures (such as FAQs in plain language) should be made easily accessible to those who wish to learn more with no hidden surprises
bull An opt-in approach to authorization should be instituted to ac-company initial DCTT rollout The feasibility and value of opt-out approaches should continue to be evaluated informed by what is technologically possible what local assessments of benefits and harms of the technology reveal over time and our evolving under-standing of the degree to which an opt-out approach is likely to increase or decrease utilization among different populations Opt-out approaches should not be precluded
Ethics of Designing and Using DCTT 69
Promoting Equity and Fairness
Digital contact tracing technologies should be designed and used in ways that as far as possible promote an equitable distribution of benefits and burdens DCTT should be deployed in a manner that does not propa-gate preexisting patterns of unfair disadvantage or distribute harms and risks unfairly throughout the population For example communities with lower rates of technology and data access may benefit less from DCTT Special attention must be paid to communities that experience preexisting health disparities and to those that are being hardest hit by the pandemic
Digital Disparities
In the United States February 2019 data indicate that approximately 80 of the population are smartphone users (Pew Research Center 2020) though rates of mobile phone use are significantly lower among people over age 65 (53) people with any disability (58 2016 data) (Anderson and Perrin 2017) people with less than a high school edu-cation (66) people who earn less than $30000 per year (71) and people who live in rural areas (71) As a result these populations and communities may use DCTT in lower numbers thereby lessening the effectiveness of DCTT and the likelihood of benefit for these populations from such systems Moreover it has been reported that many older and less costly smartphones (roughly estimated at 10ndash20 of smartphones in the US) lack important capabilities required for the leading AppleGoogle platform to work (Bradshaw 2020) This is of special concern because some of the above groups that are less likely to own smartphones in general are also less likely to own newer smartphones with the needed capabilities Some within the above groups (eg people who are older and people identified as Hispanic African American or American Indian) are also disproportionately experiencing morbidity and mortality from COVID-19 (CDC 2020h)
One may argue that by using DCTT human and financial resources that would otherwise be spent on manual contact tracing will be pre-served and these resources can then be redirected to better meet the needs of those who are not otherwise being effectively served by the technology because of disparities or for other reasons This argument has intuitive
70 Digital Contact Tracing for Pandemic Response
appeal and should be taken seriously however it is unsettled whether DCTT will contribute sufficient efficiencies to the overall public health response to make it possible financially and logistically for manual ser-vices to be allocated in greater proportion to those who are unable to benefit from DCTT It is entirely possible that at least in the short-term DCTT may introduce new inefficiencies due to unintended consequences or the need for public health officials to follow up many more contacts One possible mitigation to the challenge of digital disparitymdashthough it does not solve the underlying challenge of ensuring net efficiency across systemsmdashmight be to provide mobile phones or other devices and data packages to those who would otherwise be left out
Disparate Risk of Harm from Surveillance and Data Gathering
Ensuring wide digital coverage does not however resolve other equity concerns It is important to consider that some populations may experi-ence greater harm and fear of harm from having their data collected For example some groups such as African Americans Hispanic Americans Muslim Americans and undocumented immigrants have more reasonable fear of their data being handed over to law or immigration enforcement and some groups have lower levels of trust in public health due to past injustices (CSM 2017 Pew Research Center 2017 Rodrigues et al 2018 Auxier et al 2019) Any data gathered by DCTT should be used solely for public health purposes Efforts should be made to assure members of these and other communities that their data will not be misused or made available to those outside of a public health context In addition if DCTT are used in the current pandemic this should be with the understanding that future use of DCTT in other contexts (eg law enforcement or im-migration enforcement) is presumptively unethical
Some preliminary polling related specifically to DCTT emphasizes the complexity of the challenges faced and the need for deeper public engagement (Anderson and Auxier 2020) The polling results suggest that people who identify as African American or Hispanic are more likely than people who identify as White to consider government tracking of mobile phones as acceptable These findings like many others are difficult to in-terpret given background political polarization on the issue More direct engagement is required to better understand how different communities comprehend and experience DCTT and other forms of surveillance
Ethics of Designing and Using DCTT 71
Discrimination and Stigma
Stigma may result from an individual being identified as COV+ or a neighborhood or establishment becoming identified as a ldquohotspotrdquo as a result of numerous COV+ people living in that area or having visited that establishment In particular certain groups may suffer more as a result of being associated with COVID-19 such as the well-documented blame that has been directed toward Chinese people (and broadly East Asian communities) or the communities that are disproportionately likely to contract the illness (Devakumar et al 2020) When identifiable lo-cation data are made public as has been the case in South Korea per-sonal and private information were revealed Furthermore businesses in South Korea that were identified as having patrons who tested positive for COVID-19 have suffered economic losses and stigma (N Kim 2020)
To avoid the stigma and potential discrimination that can result from being identified as COV+ DCTT must never make data publicly avail-able that could be used to identify persons who have tested positive Safe-guards must be in place to ensure that any identifiable data that may be gathered for public health purposes are protected If DCTT data are used to provide heat maps to the public of locations that COV+ individuals frequently visit so as to provide representations of geographic risk or for other reasons it is essential that care be taken to avoid unfairly distrib-uting further economic burdens or other stigmatizing and discriminatory outcomes
Recommendations
bull A commitment to equity means a commitment to ensuring that the benefits and burdens of DCTT are distributed fairly Public engagement is an important tool for assessing impact and to rectify inequities
bull States localities and institutions that recommend widespread use of DCTT should provide technology (eg mobile phones Blue-tooth devices) and free data packages to those who desire but lack access to these devices
bull If there are lower rates of adoption of DCTT systems in some identifiable communities public health authorities should iden-
72 Digital Contact Tracing for Pandemic Response
tify ways to compensate For example directing more non-DCTT resources and efforts toward those communities to meet specific needs that are elsewhere being supported by technology
bull If maps are generated based on DCTT to provide the public with the locations that COV+ individuals have visited steps must be taken to minimize the stigma and potential financial losses that could result from being identified as a hotspot
Instituting Transparent Governance and Oversight
DCTT must be developed with an eye toward both present and future implications Transparent and publicly trustworthy management gover-nance and oversight of DCTT technology and data is both a near- and long-term necessity We face significant uncertainties DCTT technologies are rapidly developing Their risks capabilities effectiveness and down-stream implications are not yet well understood
Concerns about ldquoSurveillance Creeprdquo
Significant concerns have been expressed by privacy advocates (Guari-glia 2020) and in the popular press (Giglio 2020) about what is known as ldquosurveillance creeprdquo Their worry is that state and corporate actors will use new surveillance technologies capacities and permissions well beyond the purposes for which they were initially justified to the public and beyond the time when they are useful for the COVID-19 pandemic
Surveillance creep should be guarded against Only those data that are necessary and relevant for the public health purposes at hand should be collected and used and data should be kept only for the period of time needed for those public health purposes Data should be used only for public health purposes
Any use of DCTT during the current pandemic would be justified by the circumstances of this pandemic and its use in this context does not set a precedent for future public health use (eg use in seasonal flu surveillance efforts) Future use will require independent justification Use of DCTT in the future in other contexts (eg law enforcement or immi-gration enforcement) is presumptively unethical
Broadly speaking efforts should be made to generate public aware-
Ethics of Designing and Using DCTT 73
ness and consensus that use of DCTT in COVID-19 efforts does not imply that future use is justifiable However generating this public aware-ness may be particularly challenging given the complexity of the informa-tional environment where public debate ranges from legitimate concerns about surveillance creep to conspiracy theories regarding the origins of the COVID-19 pandemic (Muller 2020) This means authorities bear spe-cial obligations to be clear on how they plan to use the technologies what oversight mechanisms will be employed to address potential abuse and how they intend to publicize the conditions under which programs will be terminated making sure they are followed
Oversight and Ethical Review
We are rapidly gaining knowledge about SARS-CoV-2 and COVID-19 but we still have essential gaps in our understanding In the United States public health responses including DCTT will generally be developed and coordinated by individual states regional consortia and associations (Reston Sgueglia and Mossburg 2020) Good governance in this con-text requires transparency and the creation of oversight bodies with the appropriate expertise and representation to allow nimble and effective responses while serving as trusted representatives
To address the range of ethics-related concerns about the design and use of DCTT digital surveillance oversight committees should be estab-lished perhaps at a state level and with a platform for national coordina-tion These committees can provide ethical and regulatory review prior to and concurrent with widespread use of DCTT These committees should be composed of a diverse group of experts capable of evaluating a DCTT system locally including members of communities that experience higher rates of digital disparity
When assessing the design and use of digital contact tracing systems these committees (and the public more widely) should consider not only the risks and benefits accrued during the COVID-19 pandemic but also implications for the future What kind of precedent might use of these technologies during the current pandemic set for future use capabilities in other infectious disease outbreaks or in other social contexts (eg law enforcement) How can we navigate safe use of these technologies in a way that preserves public trust in them and enables the possibility of future beneficial use
74 Digital Contact Tracing for Pandemic Response
As a start it should be emphasized that principles offered in this and other guidance documents do not apply only during the pandemic Future efforts to advance DCTT capabilities during quieter times should make every effort to follow them
Recommendations
bull Digital surveillance oversight committees should be established expeditiously with diverse and qualified membership to provide ethical and regulatory review prior to and concurrent with wide-spread use of a DCTT system
bull Understandable and publicly accessible rules must guide the col-lection access control use storage and combination of data by government authorities public and private institutions and other parties such as public health researchers
bull Only those data that are necessary and relevant for the public health response to COVID-19 should be collected and used
bull Identifiable data should be kept only for the period of time needed for the public health response to COVID-19
bull Identifiable data collected as part of this response should not be shared with anyone other than the relevant public health authorities without additional specific informed consent of individual users
bull Before a government or institution adopts a digital contact tracing program they should state the conditions under which the digital contact tracing program will be terminated
bull Future use of DCTT to advance public health or other efforts (eg use in seasonal flu surveillance) would require independent justification DCTT designed for public health use should not be used by law or immigration enforcement
bull The principles offered in this guidance document apply both during and following the COVID-19 pandemic
75
The implementation of digital contact tracing technology (DCTT) is likely to implicate a number of US laws at both the federal and state levels This section focuses primarily on federal laws as these laws apply nationwide and generally preempt conflicting state laws A comprehen-sive assessment of the legality of any particular DCTT program would require case-specific analysis and attention to relevant state laws includ-ing any that specifically address DCTT which may soon exist in one or more states The analysis here is limited to the United States foreign and international laws will not be addressed
Many of the laws discussed in this section are privacy laws designed to protect individuals from the harms that may result from the unautho-rized or improper use of their personally identifiable information (PII) Under these laws legal concerns will generally be minimized if privacy protections are built directly into the DCTT technology (eg ldquoprivacy by designrdquo) As a general principle DCTT should be designed to collect and store only as much PII as is necessary to achieve the public health purpose Collecting only proximity data for example is likely to raise fewer legal concerns than collecting both proximity data and geolocation data Likewise creating aggregated anonymized or de-identified data will raise fewer legal concerns than using and disclosing PII
As we have argued elsewhere in this guidance document however the public health and societal crisis caused by COVID-19 may justify
Legal Considerations
FOUR
76 Digital Contact Tracing for Pandemic Response
greater encroachments on individual privacy than would otherwise be permissible Regardless of the type of data collected privacy concerns will be reduced if users are afforded the right to choose whether their PII is collected and how it is used and disclosed As such DCTT should gener-ally secure meaningful user consent before collecting PII a process which typically requires both disclosure of relevant information and agreement on the part of the user
Privacy concerns will also be reduced if the use of PII is strictly lim-ited to tracking and limiting the spread of SARS-CoV-2 The use of DCTT data for other purposesmdashsuch as commercial or law enforcement pur-posesmdashwould raise additional legal and ethical concerns In addition DCTT developers may be required to implement governance policies that ensure the secure storage of PII limit data retention periods require transparency about data sharing and maintain records of responses to data requests from government authorities
In short the legality of a DCTT program under current United States law will depend on a number of factors including what type of data is collected how the data are used and who may access them how user consent is obtained whether the entity collecting and using the data is the government or a private corporation the context in which data are collected (eg employment education or commercial) and which states have jurisdiction over the program
Privacy law in the United States unlike in other jurisdictions such as the European Union (EU) and Australia is generally sector-specific and limited in scope The result is a patchwork of protections that differ significantly depending on the entity that collects the data and the type of data collected For example under current law telecommunication carriers are governed by different privacy rules than mobile broadband providers Given the complexity of existing federal privacy law we be-lieve that it would be beneficial for the US Congress to enact new privacy legislation that is specifically tailored to the use of DCTT in response to COVID-19 Congress appears poised at least to debate such legislation a pair of bills recently introduced in the Senate and one in the House of Representatives would significantly restrict the collection of PII by digital devices for COVID-tracing purposes S3663 S3749 HR 6866 116th Cong (2020)
Legal Considerations 77
Data Privacy and Data Security Laws
Telecommunications
A DCTT provider that collects data from a userrsquos mobile phone may be subject to the privacy rules governing telecommunication carriers which are enforced by the Federal Communications Commission (FCC) The data protected under these rules are limited however to certain types of PII termed ldquocustomer proprietary network informationrdquo (CPNI) More-over the rules generally apply only to telecommunications carriers and interconnected VoIP (Voice over Internet Protocol) providers
In particular under section 222 of the Communications Act of 1934 47 USC sect 222 and the implementing regulations of the Federal Com-munications Commission (FCC) telecommunications carriers and VoIP providers must establish and maintain systems designed to ensure that they adequately protect their subscribersrsquo CPNI and they are generally restricted from using or disclosing CPNI without the customerrsquos consent (unless the use of disclosure is needed to provide the services subscribed to by the customer) If customer consent is sought to use or disclose CPNI individual notice must be provided to the customer and such notice must provide sufficient information to enable the customer to make an in-formed decision as to whether to permit the requested use or disclosure
CPNI is individually identifiable information that carriers and pro-viders have collected about their customers including phone numbers called and the frequency duration and timing of such calls Of most relevance to DCTT a recent FCC Notice of Apparent Liability asserted that user geolocation data collected by mobile phone network carriers qualify as CPNI under sect 222 and related rules 35 FCC Rcd 1785 (2) (2020) Pursuant to this notice the FCC fined T-Mobile for selling to third parties location data that were derived from the communication between the mobile phones of T-Mobilersquos customers and nearby network signal towers (The FCC also levied fines against ATampT Verizon and Sprint on the same grounds (Valentino-DeVries 2020)) While the FCC has made its position clear that geolocation data are CPNI courts have yet to weigh in on the matter
Even if geolocation data are CPNI however the FCC can enforce sect 222 of the Communications Act only against telecom carriers and VoIP
78 Digital Contact Tracing for Pandemic Response
providers not against cable broadband and mobile broadband internet providers 47 USC sect 53(44) 47 CFR sect 93 In 2018 the FCC promul-gated a regulation stating that contrary to its prior position its sect 222 authority does not extend to cable broadband and mobile broadband internet providers Restoring Internet Freedom 83 Fed Reg 7852 (Feb 2 2020) (to be codified at 47 CFR pts 1 8 and 20) This regulatory shift was subsequently upheld by the DC Circuit Mozilla Corporation v Federal Communications Commission 940 F3d 1 (2019)
In addition to sect 222 the FCC has authority to regulate ldquocommon carriersrdquomdashincluding both telecommunication carriers and broadband internet providersmdashunder sect 201(b) of the Communications Act In the past the FCC has interpreted sect 201(b) to protect against ldquounjust and unreasonablerdquo privacy and data security practices with respect to custom-ersrsquo personal information beyond CPNI In 2016 the FCC promulgated a regulation asserting its authority under this interpretation However Congress overturned this regulation pursuant to the Congressional Re-view Act in 2017 SJ Res 34 115th Cong (2017) At present the extent of the FCCrsquos authority under sect 201(b) remains unsettled (Mulligan and Linebaugh 2019)
Consumer Protection
The collection storage release and transmission of digital user data in-cluding proximity contacts is more generally governed by the Federal Trade Commission (FTC) The FTC is an independent US law enforce-ment agency tasked with protecting consumers and promoting competi-tion across broad sectors of the economy (FTC 2020) The FTCrsquos primary legal authority with respect to consumer protection comes from Section 5 of the FTC Act which prohibits ldquounfair or deceptive acts or practices in or affecting commercerdquo 15 USC sect 45(a)(1) Note that the FTC and FCC have some overlapping authority to protect consumer privacy in the context of telecommunications (FCC and FTC 2017)
The FTC has interpreted Section 5 to require companies to be trans-parent and accurate about their collection of PII from consumers A com-pany may be found to have engaged in a deceptive practice if it fails to disclose that it is collecting user data or fails to disclose that it is sharing these data with third parties and to provide a general description of these third parties The FTC has used its authority under Section 5 numerous
Legal Considerations 79
times to discipline companies that purport in published privacy policies or other notices to provide protection for the privacy andor security of personal information yet fail to do so in practice For example the FTC may find it both ldquounfairrdquo and ldquodeceptiverdquo for a mobile app privacy policy to state that the app never discloses location information to third parties when in fact the app shares that information with the app developerrsquos service provider which in turn uses it to provide analytical data to the app developer that are used to create targeted advertising
The FTC does not use its Section 5 authority other than to protect consumers and generally does not consider ldquode-identifiedrdquo user data which are data that are not ldquoreasonably linkablerdquo to a consumer to be a subject for consumer protection In general data collected are not ldquorea-sonably linkablerdquo so long as the company collecting it ldquo(1) takes rea-sonable measures to ensure that the data are de-identified (2) publicly commits not to try to reidentify the data and (3) contractually prohibits downstream recipients from trying to reidentify the datardquo (FTC 2012)
Many states have laws that are similar to Section 5 prohibiting un-fair and deceptive acts and practices Both Section 5 and these similar state laws can be violated not only by misrepresentations (affirmative deception) but also by material omissions Thus a failure to inform an app user of the apprsquos collection of tracking data and the planned use and disclosure of those data could constitute a violation of these laws Com-panies providing DCTT apps should make sure that all such information is disclosed in the appsrsquo terms of use to which users must affirmatively agree
Childrenrsquos Online Privacy
Children who use DCTT may be protected by additional privacy protec-tions In particular collection of digital PII from children under the age of 13 is strictly regulated under the Childrenrsquos Online Privacy Protection Act (COPPA) (15 USC sectsect 6501ndash6505) Under COPPA PII includes ldquofirst and last name[] a persistent identifier that can be used to recognize a user over time and across different online services[] and geolocation infor-mation sufficient to identify street name and name of a city or town[]rdquo COPPA prohibits a website or online service from collecting personal information (including location information) from children under age 13 without obtaining verifiable parental consent Note that there may be an
80 Digital Contact Tracing for Pandemic Response
exception to this requirement for an ldquoinvestigation on a matter related to public safetyrdquo 16 CFR sect 3125(c)(6)(iv)
Electronic Surveillance
In addition to misuse of user data by DCTT providers another privacy concern is that a third party may be able to access sensitive PII that is collected and stored by a DCTT system without the userrsquos knowledge and consent There are a number of federal criminal laws however that would likely prohibit such unauthorized access to PII
In particular the Electronic Communications Privacy Act of 1986 (ECPA)mdashwhich includes the Wiretap Act (18 USC sectsect 2510ndash2522) the Stored Communications Act (18 USC sectsect 2701ndash2711) and the Pen Register Act (18 USC sectsect 3121ndash3127)mdashmakes it a crime to access elec-tronic communications without authorization Individuals who violate the ECPA face up to five years in prison and fines up to $250000 Victims are also entitled to bring civil suits and recover actual damages in addi-tion to punitive damages and attorneyrsquos fees for violations
Generally the access restrictions in the ECPA apply unless consent is given or if access is authorized by statute for law enforcement purposes For example an employer is generally forbidden from accessing an em-ployeersquos private emails However if consent is given in the form of an employment contract that explicitly authorizes the employer to access emails it may be lawful under the ECPA for the employer to access such information Along the same lines the ECPA would likely prohibit an employer from accessing contact tracing data on an employeersquos phone without the employeersquos consent However the ECPA would likely not prohibit duly authorized government public health officials from access-ing contact tracing data without consent
As its name suggests the Stored Communications Act (SCA) regu-lates access to communications at rest that is not in transit The SCA makes it unlawful to intentionally access a facility in which electronic communication services are provided and to obtain alter or prevent au-thorized access to a wire or electronic communication while it is in elec-tronic storage in such a system As such the SCA would likely apply only to centralized collection of contact tracing data
The Pen Register Act covers any ldquosignaling informationrdquo exchanged in a communication such as phone numbers The statute does not reach
Legal Considerations 81
the content of such communications however An expansive interpreta-tion of the Pen Register Act would cover Bluetooth ldquohandshakesrdquo as they are merely signaling information between devices which do not carry content See United States v Forrester 512 F3d 500 (9th Cir 2007) (find-ing that email headers and IP addresses are akin to pen registers and have no Fourth Amendment protection) Unlike the SCA there is no statutory exclusionary rule that applies when the government illegally uses a pen register trap and trace device Additionally there is no private cause of action against the government for violations of the Pen Register Act
State Data Privacy Laws
States have a variety of privacy laws and are increasingly seeking to reg-ulate the online collection of personal information and the use and dis-closure of such information To date most of these laws focus more on transparency and protection from unauthorized access than on restricted collection and use (except with respect to biometric information) seek-ing to ensure that individuals who use websites or online services such as mobile applications do so on an informed basis with respect to the privacy provided by those sites and services Two examples of such state laws are the California Online Privacy Protection Act (CalOPPA) and the California Consumer Privacy Act (CCPA) Both laws require notice to in-dividuals who use websites or online services such as mobile applications in order to ensure that users are informed about the privacy of personal information collected by those sites and services (The CCPA also applies to data collection off-line) Both laws treat IP addresses and location data as types of potentially identifiable personal data and so would very likely apply to DCTT apps used by California residents
CalOPPA requires that the operator of any website mobile appli-cation or other online service (ldquoSiterdquo) post a privacy policy on the Site disclosing certain information regarding the Sitersquos collection use and dis-closure of PII CalOPPA applies to any Site that is accessible to California residents The required disclosures are not onerous and would apply only to collection of data that are identifiable to an individual person (but depending on who collects the data location data together with a device identifier are identifiable to the user)
The CCPA requires that any entity qualifying as a ldquobusinessrdquo provide its ldquoconsumersrdquomdashdefined as lawful residents of Californiamdashwith specific
82 Digital Contact Tracing for Pandemic Response
disclosures about the businessrsquos collection use and disclosure of personal information Importantly the CCPA applies only to for-profit businesses that meet certain thresholds of revenue or access to consumer informa-tion A public health agency or a nonprofit organization would not be subject to the CCPA Cal Civ Code sect 1798140(c)
The CCPA defines ldquopersonal informationrdquo as ldquoinformation that iden-tifies relates to describes is reasonably capable of being associated with or could reasonably be linked directly or indirectly with a particular consumer or householdrdquo The statute provides a nonexclusive list of po-tential identifiable personal information including ldquogeolocation datardquo In accordance with the CCPA businesses must provide consumers with a notice ldquoat or before the point of collectionrdquo of personal information which must describe the personal information to be collected and the pur-poses for collecting that information Businesses must additionally allow consumers to request access to and request deletion of personal informa-tion Businesses must allow for consumers to opt-out of the sale of any personal information Developers of COVID-tracing apps would want to build in compliance with these requirements In addition California Civil Code sect 1798815(a)(1) requires companies to ldquomaintain reason-able security procedures and practices appropriate to the nature of the information it processesrdquo
Like privacy laws generally the CCPA does not grant consumers rights regarding the use of de-identified information However the CCPA does require businesses to implement processes that prohibit re-identification of de-identified information as well as technical safeguards to prevent inadvertent release of that information Cal Civ Code sect 1798140(h)
Health Information Privacy
Many DCTT systems will be designed to collect health-related data of users such as symptom tracking SARS-CoV-2 test results and prior ex-posure to a person who is COV+ Individuals may have additional privacy protections with respect to the use and disclosure of this health-related information
The use and disclosure of individually identifiable health information is strictly regulated under the privacy and security rules implementing the
Legal Considerations 83
Health Insurance Portability and Accountability Act (HIPAA) HIPAA is limited in application however to health care providers and health insur-ance plans (ldquocovered entitiesrdquo) and ldquobusiness associatesrdquo of such entities ldquoBusiness associatesrdquo under HIPAA are persons who perform services for covered entities and need access to personal health information to do so
HIPAA-covered entities must have written authorization to use or disclose identifiable health information (ldquoprotected health informationrdquo or PHI) from the individual to whom such information pertains unless the HIPAA regulations promulgated by the US Department of Health and Human Services (HHS) provide an exception to the requirement for such individual authorization
Among the exceptions to the individual authorization requirement is an exception for certain uses and disclosures of PHI for public health purposes 45 CFR sect 164512(b) This exception would permit for exam-ple a HIPAA-covered entity to disclose the PHI of an individual who tests positive for SARS-CoV-2 to a public health authority A ldquopublic health authorityrdquo is an agency or authority of the US government a state ter-ritory a political subdivision of a state or territory or Indian tribe that is responsible for public health matters as part of its official mandate as well as a person or entity acting under a grant of authority from or under a contract with a public health agency such as a contact tracer Id sect 164501
Many DCTT developers are HIPAA business associates and any use and disclosure of PHI collected through DCTT used on behalf of HI-PAA-covered entities is restricted under the HIPAA privacy rules Nota-bly in response to COVID-19 HHS announced that its Office for Civil Rights would exercise its enforcement discretion and would not impose penalties for violations of certain provisions of the HIPAA Privacy Rule against health care providers or their business associates for the good faith uses and disclosures of protected health information for public health and health oversight activities during the nationwide public health emergency 85 FR 19392 (2020)
Many states also have health information privacy laws The HIPAA privacy rule sets a ldquofloorrdquo of privacy protections allowing the states to be more protective of privacy More specifically HIPAA preempts a state law if (but only if) the state law is ldquocontraryrdquo and less protective of privacy than the HIPAA privacy rule However if a state law is determined by the
84 Digital Contact Tracing for Pandemic Response
Secretary of HSS to be necessary to serve a ldquocompelling need related to public health safety or welfarerdquo it may survive preemption even if it is less privacy-protective than HIPAA 45 CFR sect 160203 (a)(1)(iv)
The Public Health Service Act also restricts the use of certain per-sonally identifiable information collected by entities involved with public health activities without the individualrsquos consent 42 USC 242m(d)
Labor and Employment Privacy Rights
Labor and employment lawsmdashthat is laws that govern the relationships between employers and employeesmdashmay prove relevant to DCTT espe-cially if employers mandate the use of DCTT or seek to collect health information regarding their employees using DCTT Depending on the built-in privacy protections of the DCTT system an employer may be able to access important health information from an employeersquos phone As noted above the ECPA would generally prohibit an employer from ac-cessing this information without the employeersquos consent Even with con-sent however there are limits on the collection and use of an employeersquos health information
In particular the use of DCTT may raise special concerns about em-ployment discrimination for example if an employer were to fire an em-ployee who tests positive for SARS-CoV-2 (COV+) or who has a known SARS-CoV-2 exposure The Americans with Disabilities Act (ADA) pro-tects disabled employees from discrimination and restricts the collection of personal health information by employers The Equal Employment Opportunity Commission (EEOC) which is the federal agency tasked with enforcing the ADA in the employment context would likely con-sider COV+ to be a ldquodisabilityrdquo under the ADA and analogous state laws prohibiting discrimination against disabled people COV+ is likely to be a ldquodisabilityrdquo especially where the individual is symptomatic andor experi-ences related health issues or if it is later determined that testing positive for SARS-CoV-2 leads to long-term or chronic health effects ldquoExposure to a COV+ personrdquo could also be covered by those laws because a person exposed to a COV+ individual could well be perceived as being disabled by being considered likely to be infected
Legal Considerations 85
The ADA generally requires that businesses make ldquoreasonable accom-modationsrdquo for persons who are disabled which may include individuals who are COV+ or who have a preexisting disability that places them at higher risk from or may be exacerbated by COVID-19 The EEOC has published guidance on reasonable accommodations under the ADA and related laws in the context of COVID-19 (EEOC 2020) Among other things this guidance clarifies that consistent with the ADA employers may take temperatures or otherwise collect health information about employees during the pandemic crisis so long as they keep that infor-mation confidential As of May 18 2020 the EEOC has not provided guidance that specifically addresses the applicability of the ADA to the use of DCTT by employers
In addition employment laws such as the ADA and the Family and Medical Leave Act (FMLA) and state law equivalents generally limit disclosure of information and require employers to keep confidential any employee personal health information related to a disability or request for medical leave Under the ADA any information regarding the medical condition or history of an employee that an employer obtains as part of an examination or inquiry into a disability could constitute a confidential medical record that can be disclosed only to certain individuals in lim-ited circumstances 42 USC sectsect 12112(d)(3)(B) and 12112(d)(4) The FMLA also prevents the disclosure of records related to medical histories in connection with an employeersquos leave request or eligibility 29 CFR sect 825500(g) The EEOC and some courts have gone further and taken the position that any information concerning an employeersquos medical con-dition is protected under the ADA or FMLA
As discussed elsewhere in this guidance document employers may have a good reason to employ DCTT in order to ensure workplace safety and limit the spread of SARS-CoV-2 in the community Employers may also face legal liability if they fail to protect employees (or customers) from potential exposure or infection In particular employers have an obligation under the Occupational Safety and Health Act to keep the workplace safe for employees In response to COVID-19 the Occupa-tional Safety and Health Administration (OSHA) has developed guid-ance on preparing the workplace (OSHA 2020) The CDC has also pre-pared guidance on healthy business operations and reducing the spread
86 Digital Contact Tracing for Pandemic Response
of SARS-CoV-2 in the workplace (CDC 2020c) Employers must strike an appropriate balance between avoiding employment discrimination and promoting workplace safety
Reflecting the need for such a balance the employee protections un-der the ADA and other employment laws are not absolute and are limited by among other things the need to protect the health and safety of other employees and the public Protection for workplace safety and health generally will justify appropriately tailored measures such as inquiries into an employeersquos personal health status or whether someone has tested positive for SARS-CoV-2 temperature checks and removal of employees from the workplace who are experiencing symptoms or have tested posi-tive and have not been cleared to return to work
Note finally that the use of DCTT by employers should be evaluated in conjunction with the hazard pay sick leave and other benefits that are available to employees Under the Families First Coronavirus Response Act employers with more than 50 employees and fewer than 500 employ-ees are required to provide two weeks of paid sick leave to an employee who stays home because of COVID-19 Pub L No 116-127 134 Stat 178 (2020) This paid leave extends to those who are themselves ill are quarantined or are awaiting a diagnosis as well as those who are caring for sick family members However reporting suggests that more than 75 of US workers will not qualify for benefits under this act (Cochrane Miller and Tankersley 2020)
Constitutional Privacy Rights
A DCTT program involving only private actors operating on the ba-sis of voluntarily provided information would not present constitutional privacy issues But any government-directed use of digital technology to support public health tracking and contact tracing involving mandatory government surveillance may potentially implicate a variety of consti-tutional protections These constitutional protections apply to actions taken by any level of government in the United States While state gov-ernments have broad policing powers in the area of public health (Jacob-son v Massachusetts 197 US 11 (1905)) and are generally allowed to enforce legislation not preempted by federal laws even emergency and
Legal Considerations 87
health-protective laws must be consistent with the US Constitution (HHS 2019 CDC 2020f)
Fourth Amendment Search and Seizure
Many people considering whether to use a DCTT app may be concerned that government enforcement agencies would obtain tracing data and use those data to conduct criminal prosecutions or immigration proceed-ings Constitutional protections notably the Fourth Amendmentrsquos limit on warrantless searches limit the governmentrsquos use of personal data in the criminal context However exceptions exist allowing law enforce-ment to access information even when such access would generally be prohibited How the government accesses personal data stemming from contact tracing needs to be scrutinized and protections will hinge on the manner of access
In general the Fourth Amendment protects ldquo[t]he right of the peo-ple to be secure in their persons houses papers and effects against un-reasonable searches and seizuresrdquo As originally interpreted the Fourth Amendment was considered tied to common-law trespass That is no lon-ger the case US Supreme Court precedent interprets the Fourth Amend-ment to protect ldquopeople not placesrdquo and extends to the protection of certain expectations of privacy such as location information as long as such expectations are reasonable Katz v United States 389 US 347 351 (1967) A warrantless government search is unconstitutional when the information sought is private and such expectation of privacy is ldquoone that society is prepared to recognize as reasonablerdquo Smith v Maryland 442 US 735 743ndash44 (1979)
The constitutionality of a search will revolve around the following analysis whether the digital program either violates an individualrsquos ldquorea-sonable expectation of privacyrdquo (likely triggered by programs collecting large amounts of location andor health data) or involves a government ldquotrespassrdquo (likely triggered by required app downloads) Katz v United States 389 US 347 (1967) United States v Jones 565 US 400 (2012)
Courts will most likely weigh the intrusiveness of the measures taken in implementing a search standard against the severity of the situation governmental and individual interests and accountability measures and safeguards built into the system
Voluntary sharing by individuals of their information with other par-
88 Digital Contact Tracing for Pandemic Response
ties including the government would mean that there was no reasonable expectation of privacy and would not raise the issues elaborated above It is worth noting that consent may not be considered voluntary if coerced or conditioned especially with regard to public employees or students of public institutions
Third-Party Doctrine
Some legal doctrines allow for the governmentrsquos acquisition of otherwise private information consistent with Fourth Amendment privacy protec-tions The third-party doctrine for example provides that individuals have no reasonable expectation of privacy in information voluntarily shared with others even if the information is revealed on the assumption that it will be used only for a limited purpose and the confidence placed in the third party will not be betrayed Smith v Maryland 442 US 735 (1979) United States v Miller 425 US 435 (1976) This applies to in-formation provided by third parties (mobile carriers internet service pro-viders medical tracking device manufacturers etc) to the government under order or request even when the third partyrsquos end-user agreements or privacy policies create an expectation of privacy
The Supreme Court has narrowed the applicability of the third-party doctrine to exclude use and disclosure of ldquohistoricalrdquo cell-site location information (CSLI) data For example in Carpenter v United States 138 S Ct 2206 (2018) the Court reasoned that the third-party doctrine does not justify use and disclosure of historical CSLI because an individual does not provide that information voluntarily Rather that information is pervasively collected by the cell phone company without any affirmative action on the part of the individual The Court did not express a view on ldquoreal-timerdquo CSLImdashlocation information that live-tracks a cell phonersquos locationmdashor on GPS data that may be stored in the phone itself
The Carpenter decision builds on a line of cases related to searches of digitally stored location data In Riley v California 134 S Ct 2473 (2014) the Court held that absent exigent circumstances law enforce-ment must obtain a warrant to search an individualrsquos phone Exigent circumstances are those that require immediate action because there is a probability that evidence may be destroyed The use of a centralized data-base for collection of digital contact tracing data would obviate deletion
Legal Considerations 89
concerns If the data are stored locally in the phone issues may arise as to whether law enforcement may suspect the data may be deleted following an arrest
Similarly in United States v Jones 132 S Ct 945 (2012) Justice So-nia Sotomayor authored a concurring opinion arguing that the use of a GPS to track a defendantrsquos whereabouts has the potential of providing the government with enough data points to create a ldquomosaicrdquo of the personrsquos life Location data obtained through centralized location contact tracing have the potential of providing information on an individualrsquos where-abouts beyond whatrsquos necessary for determining proximity to infected individuals Localized data may also raise the same issues if accessed by law enforcement
Following Carpenter several courts have addressed the constitution-ality of novel location tracking In Massachusetts for instance a federal district court concluded that police use of a ldquopole camerardquo on a utility pole to investigate the movements of an individual constituted a search under the Fourth Amendment United States v Moore-Bush 381 FSupp3d 139 (D Mass 2019) The court reasoned that even in a public space an in-dividual still retains a reasonable expectation of privacy ldquoin the whole of their physical movementsrdquo Citing Carpenter and Jones the court stated that the governmentrsquos unrestrained power to collect data that reveal pri-vate aspects of identity is susceptible to abuse and gives police access to a category of information that is ldquootherwise unknowablerdquo Long-term monitoring of a personrsquos movements consequently violates that individ-ualrsquos expectation of privacy Notably the court emphasized the capability of the camera to create a searchable digital log of the photos taken for the eight-month period during which the camera was used
State courts have also weighed in on the issue The Massachusetts Supreme Judicial Court found that police access to real-time location data pinpointing an individualrsquos movement whether from a third party or a cell-site simulator infringes upon an individualrsquos reasonable expec-tation of privacy Commonwealth v Almonor 120 NE3d 1183 1195 (Mass 2019) The Washington Supreme Court for its part held that a cell phone ping used to locate the defendantrsquos vehicle in real time is a search under the Fourth Amendment requiring a warrant absent exigent circumstances State v Muhammad 428 P3d 1177 (2018) And the Colo-
90 Digital Contact Tracing for Pandemic Response
rado Court of Appeals held that police use of a video pole camera to con-tinuously surveil a defendantrsquos fenced-in backyard constitutes a search under the Fourth Amendment People v Tafoya 2019 BL 457321 Colo Ct App 17CA1243 (2019)
Application of Carpenter by lower courts to novel location-tracking tactics is still evolving and it is as yet unclear how the narrower interpre-tation of the third-party doctrine will continue to be expanded and ap-plied particularly in cases of short-term monitoring of massive amounts of location andor health data Moreover it is unclear whether Carpenter would apply to DCTT data collected by the government itself
Special Needs Doctrine
An argument in favor of the constitutionality of government DCTT programs is that the ldquospecial needsrdquo doctrine would apply Under this doctrine a warrantless search that would otherwise violate the Fourth Amendment might be permissible based on a special need relating to pub-lic health When the search is conducted for a nontraditional law enforce-ment purpose and circumstances make securing a warrant impracticable the Supreme Court has ruled that warrantless searches may be permissi-ble The special needs doctrine however is highly controversial because it is not a consistently applied Fourth Amendment exception so it is diffi-cult to predict when courts would authorize nontraditional surveillance Some factors considered by the court are (1) the balance between the intrusiveness of the government action and the anticipated public bene-fits (2) the existence of legislative authorization (3) judicial process or the ability of the subject individual to challenge the government action (4) the scope or breadth of government action and (5) the likelihood of the collected data being used in criminal proceedings The Supreme Court did note in Chandler v Miller 520 US 305 (1997) that a ldquorisk to public safety [that] is substantial and realrdquo may justify ldquoblanket suspicionless searches calibrated to the riskrdquo citing as examples the routine searches conducted at airports and entrances to some official buildings (Searches within the context of immigration are further analyzed below)
Immigration Enforcement
Exceptions apply to the constitutional requirement that a warrant ac-company an unreasonable search or seizure in the immigration context
Legal Considerations 91
For example an exception to the general warrant requirement is the bor-der search exception which allows government officials to search and seize without a warrant persons and property at the border or at the functional equivalent of a border See United States v Montoya de Her-nandez 473 US 531 (1985) United States v Flores-Montano 541 US 149 (2004) Federal regulation authorizes immigration officials to oper-ate within 100 miles of any US external boundary (See 8 CFR sect 2871 defining ldquoreasonable distancerdquo as ldquowithin 100 air miles from any external boundary of the United Statesrdquo) A functional equivalent of a border may include any airport where international flights may be received automo-bile checkpoints servicing international traffic and vessels in territorial waters Government officials however must still have ldquoreasonable suspi-cionrdquo of an immigration violation or a crime to search or seize persons or property
In the context of digital data Customs and Border Protection (CBP) officials may conduct either manual or forensic searches of electronic devices at the border or its functional equivalent A manual search is considered a routine search and may include accessing the phone and ldquobrowsingrdquo its contents If the electronic device is password protected individuals must provide information for unlocking the device Forensic searches on the other hand are nonroutine and involve a more invasive search of the electronic devicersquos contents Federal circuit courts are split on whether a CBP agent needs ldquoreasonable suspicionrdquo before conducting a forensic search of an electronic device But Supreme Court precedent clearly states that suspicionless searches are not unconstitutional when public safety is considered Skinner v Ry Labor Execsrsquo Assrsquon 489 US 602 (1989)
A recent CBP directive provides guidance and standard operating procedures regarding forensic searches of electronic devices CBP 3340-049A Border Search of Elec Devices (DHS 2018) The directive states that CBP officers may detain electronic devices or copies of the informa-tion contained within these devices for a reasonable period time not to exceed five days This directive raises the concern that travelers may be required to turn over contact tracing data stored on their phone to CBP officers Note that the directive has been challenged in federal court and is currently awaiting appeal Alasaad v Nielsen 419 FSupp3d 142 (D Mass 2019)
92 Digital Contact Tracing for Pandemic Response
Searches in Schools
Another exception to the general warrant requirement applies to searches by non-law-enforcement government officials in public schools (ie school officials) Within this context school officials have broad powers to conduct searches as long as those searches are reasonable Searches by individuals in private schools are not governed by the Fourth Amend-ment State regulation of searches in private schools varies (See US DOE 2009)
Related Federal Privacy Statutes
Outside the Fourth Amendment context certain laws provide protections against government collection of and access to personal data The USA Freedom Act of 2015 for example bans the governmentrsquos bulk collec-tion of internet metadata and telephonic records which was previously allowed under Section 215 of the USA Patriot Act The government must now identify with specificity the identity of a person account address or personal device when requesting records The law allows for the acqui-sition of data by two degrees of separationmdashor ldquohopsrdquomdashfrom targeted individuals If a centralized system in contact tracing is used it is unclear whether the government may need to resort to this provision since it would likely have consent from individuals to collect and use the data
The Privacy Act of 1974 also regulates the collection use and disclo-sure of personal data but applies only to federal agencies (and their con-tractors) not to state or local agencies 5 USC sect 552a The Act protects against disclosure of individually identifying ldquorecord[s]rdquo that are kept within a ldquosystem of recordsrdquo The Act limits disclosure of information ldquoexcept pursuant to a written request by or with prior written consent of the individual to whom the record pertainsrdquo Certain disclosures are ex-empt from the Actrsquos applicability Pertinent disclosure exceptions are for records required to be disclosed under the Freedom of Information Act (FOIA) or disclosures ldquoto a person pursuant to a showing of compelling circumstances affecting the health or safety of an individualrdquo A disclosure under FOIA however would not include information in ldquopersonnel and medical files and similar filesrdquo when disclosure ldquowould constitute a clearly unwarranted invasion of personal privacyrdquo FOIA Guide 2004 Edition
Legal Considerations 93
Exemption 6 If non-anonymized data are turned over to the federal or state governments it is important to consider whether PII would be pro-tected from disclosure under FOIA or state freedom of information laws
Consent
User consent is a cross-cutting issue for evaluating many of the laws and regulations governing personal information privacy discussed in the prior sections In general privacy laws can be justified on the grounds that an individual should have the option to control with various types and de-grees of limitation the collection use retention andor disclosure of in-formation pertaining to that individual by others As such many privacy laws start from the premise that absent an individualrsquos consent use or disclosure of that individualrsquos PII is impermissible except for certain enu-merated purposes deemed to outweigh the individualrsquos privacy interests
Consent like agreements in general can be manifest in different ways in specific circumstances In some cases an affirmative actionmdashsuch as a signaturemdashis needed to demonstrate consent In other cases inactionmdashsuch as declining to ldquounsubscriberdquo from receiving certain unsolicited emailsmdashconstitutes consent Where a law requires a ldquowrittenrdquo signature in all but a few contexts the signature may be executed electronically In the United States that means the ldquosignaturerdquo may consist of any of the following ldquoan electronic sound symbol or processrdquo so long as it is ldquoattached to or logically associated with a contract or other record and executed or adopted by a person with the intent to sign the recordrdquo Electronic Signatures in Global and National Commerce Act (E-SIGN) (15 USC 7006)
The scope of a consent depends on what was deemed to be under-stood by the consenting party That is least clear when the consent is in-ferred from inactivity even if terms stating the consequences of inactivity have been provided The scope of consent is most clear when the terms agreed to have been presented to or provided by the consenting party in a conspicuous documented manner and a record exists of those terms Courts uphold the validity of clickthrough agreements because users are deemed to review the terms to which they respond by clicking ldquoI agreerdquo
94 Digital Contact Tracing for Pandemic Response
But where terms are ambiguous or confusing buried in other text or presented obscurely the ldquoI agreerdquo action may not mean the user actually agreed to specific terms
Terms of Use and Privacy Policies for apps are often written in com-plicated or nuanced language with key points difficult to discern More-over they are generally hard to read on a mobile device Many users of mobile phone apps agree to such terms without even attempting to read or to understand them As such it is often questionable whether an app user has knowingly agreed to all the terms of those documents Presen-tation of terms in large typeface short sentences simple language and direct disclosures makes user consent more meaningful
For contact tracing apps that collect PII andor PHI consent will overcome the restrictions of many if not most privacy laws provided the consent is freely given reflects a full understanding of the terms for use collection and disclosure of the information and is confirmed by an af-firmative act such as a click that may be executed only upon a complete reading of Terms of Use and Privacy Policies Whether consent may be deemed ldquofreely givenrdquo in certain circumstances depends on contextual understandings of party relationships including the employer-employee and government-citizen relationships
Anti-discrimination and Individual Freedom Laws
Any measure taken to protect public health and safety must comply with the Constitution and civil rights laws such as the ADA that prohibit discrimination against persons in certain protected categories such as race gender religion or disability In addition certain implementations of DCTT could be challenged under a variety of individual freedom protections
Anti-discrimination Laws
In general it would be impermissible to use DCTT in a way that either targets or excludes people on the basis of their membership in one of these protected categories
When motivated by animus against a protected class as defined by law and not narrowly tailored to advance a compelling government inter-
Legal Considerations 95
est a discriminatory regulation would be considered unconstitutional un-der the Equal Protection Clause of the Fourteenth Amendment to the US Constitution See Jew Ho v Williamson 103 F10 CCND Cal (1900) (striking down a quarantine imposed by San Francisco in response to an outbreak of bubonic plague because it was racially motivated) see also Church of Lukumi Babalu Aye v Hialeah 508 US 520 (1993) (supposed public health measure unconstitutional because it targeted the practices of one religion)
The risk of unintentional yet illegal discrimination in using DCTT is a real possibility Recent studies of infection rates among the population have revealed a larger-than-proportional infection rate among certain minority communities such as Latinx African American and American Indian communities (NYC DOH 2020) Programs that target specific ra-cial ethnic tribal or religious groups may raise constitutional and other legal concerns
Religious Freedom Laws
The use of DCTT may also raise concerns about religious freedom For example there may be religious objections to restrictions on gathering for worship carrying a mobile phone or the use of imaging technology Under current Supreme Court precedent generally applicable laws that do not discriminate against religion on their face do not violate the Free Exercise Clause of the First Amendment even if those laws have an inci-dental effect on the exercise of religion Employment Div Dept of Hu-man Resources of Oregon v Smith 494 US 872 (1990) These laws need not be justified by compelling government interest (the ldquostrict scrutinyrdquo standard of review) the government need only show that they are ratio-nally related to a legitimate interest On the other hand laws that are not neutral and not of general applicability must be justified by compelling government interest and must be narrowly tailored to advance that in-terest if it burdens religious practicesmdasha very tough hurdle to overcome Church of Lukumi Babalu Aye v Hialeah 508 US 520 (1993)
This general approach however is disrupted in some contexts by statutes adopted to provide greater protection to religious freedom The federal Religious Freedom Restoration Act (RFRA) requires strict scru-tiny for federal actions that burden religion and many states have ad-opted ldquostate RFRAsrdquo that do the same for actions by state and local
96 Digital Contact Tracing for Pandemic Response
governments The Religious Land Use and Institutionalized Persons Act which extends similar protections to persons confined to an institution such as a prison jail or mental health facility may also be relevant 42 USC sect 2000cc
Under either standard of review courts will examine whether a gov-ernment action imposes a substantial burden on religious exercise if not no religious freedom violation has occurred Such a finding is unlikely for DCTT programs absent evidence that the government is using the digital information to take action against religious persons that is not necessary to avoid the spread of a serious disease Nevertheless legal challenges on religious freedom grounds cannot be ruled out
Legislative Recommendations
bull Congress should enact new legislation specifically tailored to fa-cilitate the use of DCTT as part of the public health response to COVID-19 while also protecting user privacy and ensuring data security
bull Congress should require DCTT developers to disclose to users in clear language the nature of the information that would be col-lected how it would be collected how it would be stored and for what purposes it may be used
bull While the rollout of DCTT should initially employ an opt-in au-thorization approach the feasibility acceptability and value of opt-out approaches should continue to be evaluated As such opt-out approaches to consent should not be precluded by legislation
bull Congress should prohibit the commercial use of data collected for COVID-19 response by DCTT
bull Congress should prohibit discrimination on the basis of data col-lected by DCTT
bull If Congress is unable to enact suitable legislation state legislatures should work toward enacting similar laws for their jurisdictions A ldquomodelrdquo state law should be rapidly developed to facilitate na-tionwide uniformity of legal requirements
97
Public Health
Characteristics That Make Data Useful to
Public Health for Reducing Disease Transmission
bull Technologies or apps with the goal of enhancing public health ca-pacity to identify cases and trace contacts in order to control the spread of SARS-CoV-2 should be designed to match functionality with that goal
bull Technologies or apps may produce some false negatives or false positives but they should be accurate enough that public health authorities feel confident that they support and donrsquot detract from contact tracing efforts
bull DCTT approaches for public health should be designed to facili-tate the following
deg Identifying contacts including those who may not be easily found otherwise
deg Finding and notifying contacts rapidly before they develop symptoms if infected
deg Analyzing the nature of contact to determine whether con-tact is high medium or low risk to support decisions about whether quarantine should be mandatory should be volun-tary or is not needed
deg Following up with cases and contacts so that public health can provide resources to support isolation and quarantine at home
Recommendations
FIVE
98 Digital Contact Tracing for Pandemic Response
bull Data collected through DCTT should be made available to public health professionals and to researchers in de-identified form to support population-level epidemiologic analysis
Ethics
Collecting Data to Inform Policy and Practice
bull Reviews of DCTT systems must be conducted in part by an inde-pendent intermediary that has established the publicrsquos trust
bull Those who authorize use of DCTT within a particular jurisdic-tion or institution should continuously and systematically monitor the technologyrsquos performance in that context This should include monitoring for effectiveness and benefit monitoring for harms and monitoring for the fair distribution of both benefits and harms They should also monitor evidence that is being generated in other contexts about their selected technological solution and about other competing technologies
bull Data should be available to users that would permit them to further investigate their personal risk with public health officials or other health workers to add a layer of protection against unnecessary quarantine
Public Trust and Public Attitudes
bull More research into public attitudes is needed In particular in-depth qualitative research should examine public attitudes about perceptions of trust in DCTT among different communities which features of DCTT influence trust and the extent to which people are willing to provide different types of data through DCTT to help their community
bull States and localities that are considering adopting DCTT should engage with the public to increase their understanding of the acceptability of DCTT design features and uses among diverse communities
Recommendations 99
Designing Flexible Technology to Maximize
Public Health Utility While Respecting Other Values
bull Technology companies should not alone control the terms condi-tions or capabilities of DCTT nor should they presume to know what may be acceptable to members of the public
bull A ldquovalues in designrdquo approach to development of DCTT should be adopted (Flanagan Howe and Nissenbaum 2008 Knobel and Bowker 2011) Robust public- and user-engagement activities should be pursued to identify and incorporate to the extent pos-sible a range of values into the design of the technology These values may include privacy but also autonomy efficiency equity or others Technology design should reflect an appropriate balance and prioritization of identified values
bull Technology design should not be static but should be capable of evolving depending upon local conditions new evidence and changing preferences and priorities
bull DCTT should be designed to have a base set of features that pro-tect privacy with layers of additional capabilities that users may choose to activate An initial default should be that user location data are not shared but users should be provided with easy mech-anisms and prompts to allow for opting-in to this capability with encouragement to the public if and as it is shown to be critical to achieving public health goals
Policy Positions to Advance Widespread
Use of Digital Contact Tracing Technologies
bull DCTT use should not be mandated at this time given uncertainty about potential harms and benefits Additional technology user and real-world testing is needed
bull Incentives can be a useful complement to encouragements how-ever any incentives for users to install and use DCTT must be equitable should not be coercive and should align with effective use of the technology
100 Digital Contact Tracing for Pandemic Response
bull Trusted leaders should be enlisted to communicate effectively with the public about DCTT and encourage its use should the technol-ogy demonstrate some potential The limits of knowledge regard-ing effectiveness should also be explained along with what will be done to improve technological capabilities as understanding evolves
Disclosure and AuthorizationConsent
bull A clear and concise module consisting of basic disclosure and vol-untary authorization should be developed to accompany DCTT This module should not take the form of ldquoclickwraprdquo terms of service or end-user agreements but rather provide only essential information necessary for an individual to make a decision More detailed disclosures (such as FAQs in plain language) should be made easily accessible to those who wish to learn more with no hidden surprises
bull An opt-in approach to authorization should be instituted to ac-company initial DCTT rollout The feasibility and value of opt-out approaches should continue to be evaluated informed by what is technologically possible what local assessments of benefits and harms of the technology reveal over time and our evolving un-derstanding of the degree to which an opt-out approach is likely to increase or decrease utilization among different populations Opt-out approaches should not be precluded
Promoting Equity and Fairness
bull A commitment to equity means a commitment to ensuring that the benefits and burdens of DCTT are distributed fairly Public engagement is an important tool for assessing impact and to rectify inequities
bull States localities and institutions that recommend widespread use of DCTT should provide technology (eg mobile phones Blue-tooth devices) and free data packages to those who desire but lack access to these devices
Recommendations 101
bull If there are lower rates of adoption of DCTT systems in some identifiable communities public health authorities should iden-tify ways to compensate For example directing more non-DCTT resources and efforts toward those communities to meet specific needs that are elsewhere being supported by technology
bull If maps are generated based on DCTT to provide the public with the locations that COV+ individuals have visited steps must be taken to minimize the stigma and potential financial losses that could result from being identified as a hotspot
Instituting Transparent Governance and Oversight
bull Digital surveillance oversight committees should be established expeditiously with diverse and qualified membership to provide ethical and regulatory review prior to and concurrent with wide-spread use of a DCTT system
bull Understandable and publicly accessible rules must guide the col-lection access control use storage and combination of data by government authorities public and private institutions and other parties such as public health researchers
bull Only those data that are necessary and relevant for the public health response to COVID-19 should be collected and used
bull Identifiable data should be kept only for the period of time needed for the public health response to COVID-19
bull Identifiable data collected as part of this response should not be shared with anyone other than the relevant public health authorities without additional specific informed consent of individual users
bull Before a government or institution adopts a digital contact tracing program they should state the conditions under which the digital contact tracing program will be terminated
bull Future use of DCTT to advance public health or other efforts (eg use in seasonal flu surveillance) would require independent justification DCTT designed for public health use should not be used by law or immigration enforcement
102 Digital Contact Tracing for Pandemic Response
bull The principles offered in this guidance document apply both during and following the COVID-19 pandemic
Legislative
bull Congress should enact new legislation specifically tailored to fa-cilitate the use of DCTT as part of the public health response to COVID-19 while also protecting user privacy and ensuring data security
bull Congress should require DCTT developers to disclose to users in clear language the nature of the information that would be col-lected how it would be collected how it would be stored and for what purposes it may be used
bull While the rollout of DCTT should initially employ an opt-in au-thorization approach the feasibility acceptability and value of opt-out approaches should continue to be evaluated As such opt-out approaches to consent should not be precluded by legislation
bull Congress should prohibit the commercial use of data collected for COVID-19 response by DCTT
bull Congress should prohibit discrimination on the basis of data col-lected by DCTT
bull If Congress is unable to enact suitable legislation state legislatures should work toward enacting similar laws for their jurisdictions A ldquomodelrdquo state law should be rapidly developed to facilitate na-tionwide uniformity of legal requirements
Resources
103
US Government Response
White House
The White House and CDC 16 April 2020 ldquoOpening Up America Againrdquo The White House and the Centers for Disease Control and Prevention Available at httpswwwwhitehousegovopeningamerica
Congress
A Bill to Protect the Privacy of Consumersrsquo Personal Health Information
Proximity Data Device Data and Geolocation Data during the Corona-
virus Public Health Crisis S3663 116th Cong (2019-2020) Available at httpswwwcongressgovbill116th-congresssenate-bill3663
A Bill to Protect the Privacy of Health Information during a National
Health Emergency S3749 116th Cong (2019-2020) Available at httpswwwcongressgovbill116th-congresssenate-bill3749
To Protect the Privacy of Health Information during a National Health
Emergency HR6866 116th Cong (2019-2020) Available at httpswwwcongressgovbill116th-congresshouse-bill6866
Congressional Research Service
Foster Michael 16 April 2020 ldquoCOVID-19 Digital Surveillance and Privacy Fourth Amendment Considerationsrdquo Legal Sidebar LSB10449 Congres-sional Research Service httpscrsreportscongressgovproductpdfLSBLSB10449
Mulligan Stephen P and Chris D Linebaugh 25 March 2019 ldquoData Pro-tection Law An Overviewrdquo R45631 Congressional Research Service httpscrsreportscongressgovproductpdfRR45631
104 Resources
US Department of Health and Human Services (HHS) Centers for Disease Control and Prevention (CDC)
CDC 6 May 2020 ldquoInterim Guidance for Businesses and Employersrdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention Available at httpswwwcdcgovcoronavirus2019-ncov communityguidance-business-responsehtml
CDC 3 May 2020 ldquoDiscontinuation of Isolation for Persons with COVID-19 Not in Healthcare Settingsrdquo Interim Guidance Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention httpswwwcdc govcoronavirus2019-ncovhcpdisposition-in-home-patientshtml
CDC 30 April 2020 ldquoContact Tracingrdquo Get and Keep America Open Sup-porting States Tribes Localities and Territories Centers for Disease Control and Prevention httpswwwcdcgovcoronavirus2019-ncovphpopen-americacontact-tracinghtml
CDC 29 April 2020 ldquoInterim Guidelines for Collecting Handling and Testing Clinical Specimens from Persons for Coronavirus Disease 2019 (COVID-19)rdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention Available at httpswwwcdcgovcorona-virus2019-ncovlabguidelines-clinical-specimenshtml
CDC 29 April 2020 ldquoContact Tracing Part of a Multipronged Approach to Fight the COVID-19 Pandemicrdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention httpswwwcdcgov coronavirus2019-ncovphpprinciples-contact-tracinghtml
CDC 28 April 2020 ldquoPreliminary Criteria for the Evaluation of Digital Contact Tracing Tools for COVID-19rdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention Available at httpswwwcdc govcoronavirus2019-ncovdownloadsphpprelim-eval-criteria-digital- contact-tracingpdf
CDC 20 April 2020 ldquoDigital Contact Tracing Tools for COVID-19rdquo Corona- virus Disease 2019 (COVID-19) Centers for Disease Control and Preven-tion Available at httpswwwcdcgovcoronavirus2019-ncovdownloadsdigital-contact-tracingpdf
CDC 6 April 2020 ldquoOperational Considerations for the Identification of Healthcare Workers and Inpatients with Suspected COVID-19 in Non-US Healthcare Settingsrdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention Available at httpswwwcdcgovcoron virus2019-ncovhcpnon-us-settingsguidance-identify-hcw-patientshtml
HHS 2 April 2020 ldquoNotification of Enforcement Discretion under HIPAA to Allow Uses and Disclosures of Protected Health Information by Business
Resources 105
Associates for Public Health and Health Oversight Activities in Response to COVID-19rdquo 45 CFR Parts 160 and 164 Available at httpswwwhhsgovsitesdefaultfilesnotification-enforcement-discretion-hipaapdf
HHS 30 March 2020 Notification of Enforcement Discretion for Telehealth Remote Communications during the COVID-19 Nationwide Public Health Emergency US Department of Health and Human Services Office of Civil Rights Available at httpswwwhhsgovhipaafor-professionalsspecial- topicsemergency-preparednessnotification-enforcement-discretion- telehealthindexhtml
CDC 24 February 2020 ldquoLegal Authorities for Isolation and Quarantinerdquo Centers for Disease Control and Prevention Available at httpswww cdcgovquarantineaboutlawsregulationsquarantineisolationhtml
HHS 26 November 2019 ldquoPublic Health Emergency Declarationrdquo Public Health Emergency US Department of Health and Human Services Office of the Assistant Secretary for Preparedness and Response Available at httpswwwphegovPreparednesslegalPagesphedeclarationaspx
CDC September 2018 ldquoPublic Health Surveillance Preparing for the Futurerdquo Centers for Disease Control and Prevention Available at httpswww cdcgovsurveillancepdfsSurveillance-Series-Booklethpdf
CDC 2005 ldquoVI Plan for Surveillance of Contacts of SARS Case Supplement B SARS Surveillancerdquo Centers for Disease Control and Prevention Available at httpswwwcdcgovsarsguidanceb-surveillancecontactshtml
National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research 1979 ldquoThe Belmont Report Ethical Principles and Guidelines for the Protection of Human Subjects of Researchrdquo US Depart-ment of Health and Human Services Office of Human Research Protec-tions httpswwwhhsgovohrpregulations-and-policybelmont-reportread-the-belmont-reportindexhtml
US Equal Employment Opportunity Commission (EEOC)
EEOC 7 May 2020 ldquoWhat You Should Know About COVID-19 and the ADA the Rehabilitation Act and Other EEO Lawsrdquo US Equal Employ-ment Opportunity Commission Available at httpswwweeocgovwyskwhat-you-should-know-about-covid-19-and-ada-rehabilitation-act-and-other-eeo-laws
EEOC 21 March 2020 ldquoPandemic Preparedness in the Workplace and the Americans with Disabilities Actrdquo US Equal Employment Oppor-tunity Commission Available at httpswwweeocgovlawsguidancepandemic-preparedness-workplace-and-americans-disabilities-act
106 Resources
Federal Trade Commission (FTC)
FTC 2020 ldquoPrivacy amp Data Security Update 2019rdquo Federal Trade Commis-sion httpswwwftcgovsystemfilesdocumentsreportsprivacy-data- security-update-20192019-privacy-data-security-report-508pdf
FTC 2012 ldquoProtecting Consumer Privacy in an Era of Rapid Changerdquo FTC Report Recommendations for Businesses and Policymakers Fed-eral Trade Commission httpswwwftcgovsitesdefaultfilesdocumentsreportsfederal-trade-commission-report-protecting-consumer-privacy-era- rapid-change-recommendations120326privacyreportpdf
Occupational Safety and Health Administration (OSHA)
OSHA April 2020 ldquoGuidance on Preparing Workplaces for COVID-19rdquo OSHA 3990-03 2020 Department of Labor Occupational Safety and Health Act of 1970 Available at httpswwwoshagovPublicationsOSHA3990pdf
Customs and Border Protection (CBP)
CBP 4 January 2018 ldquoBorder Search of Electronic Mediardquo US Customs and Border Protection CDP DIRECTIVE NO 3340-049A Available at httpswwwcbpgovsitesdefaultfilesassetsdocuments2018-JanCBP-Directive-3340-049A-Border-Search-of-Electronic-Media-Compliantpdf
Other Governmental and Nongovernmental Organizations
Contact Tracing Surveillance PlansMethods
Simmons-Duffin Selena 7 May 2020 ldquoStates Nearly Doubled Plans for Contact Tracers since NPR Surveyed Them 10 Days Agordquo NPR Available at httpswwwnprorgsectionshealth-shots20200428846736937we-asked-all-50-states-about-their-contact-tracing-capacity-heres-what- we-learned
Resolve to Save Lives 2020 ldquoCOVID-19 Contact Tracing Playbookrdquo Vital Strategies Available at httpscontacttracingplaybookresolvetosavelives org
Simpson Erin and Adam Conner 22 April 2020 ldquoDigital Contact Tracing to Contain the Coronavirusrdquo Center for American Progress Avail-able at httpswwwamericanprogressorgissuestechnology-policynews20200422483521digital-contact-tracing-contain-coronavirus
Watson Crystal Anita Cicero James Blumenstock and Michael Fraser 10 April 2020 ldquoA National Plan to Enable Comprehensive COVID-19 Case Finding and Contact Tracing in the USrdquo Johns Hopkins Bloomberg School of Pub-
Resources 107
lic Health Center for Health Security and the Association of State and Ter-ritorial Health Officials Available at httpswwwcenterforhealthsecurity orgour-workpubs_archivepubs-pdfs2020200410-national-plan-to- contact-tracingpdf
Africa CDC 9 April 2020 ldquoGuidance on Contact Tracing for COVID-19 Pandemicrdquo Manuals Guidelines amp Frameworks African Union Africa CDC Available at httpsafricacdcorgdownloadguidance-on-contact- tracing-for-covid-19-pandemic
PIH 4 April 2020 ldquoPart I Testing Contact Tracing and Community Manage-ment of COVID-19rdquo PIH Guide | COVID-19 STOP COVID Partners in Health Available at httpswwwpihorgsitesdefaultfiles2020-04PIH_Guide_COVID_Part_I_Testing_Tracing_Community_Managment_4_4pdf
Hellewell Joel Sam Abbott Amy Gimma et al 28 February 2020 ldquoFea- sibility of Controlling COVID-19 Outbreaks by Isolation of Cases and Contactsrdquo The Lancet Global Health 8 e488ndash96 Available at https doiorg101016S2214-109X(20)30074-7
Contact Tracing EthicsPrinciplesGuidance
Center for Democracy amp Technology (CDT)ndashhttpscdtorginsightsCDT 30 April 2020 ldquoStatement of the Center for Democracy amp Technol-
ogy Regarding Use of Data to Fight COVID-19rdquo Center for Democ-racy amp Technology httpscdtorgwp-contentuploads202004 CDT-Statement-Regarding-Use-of-Data-to-Fight-COVID-19pdf
Shetty Ridhi 23 April 2020 ldquoData Use in the Fight against COVID-19 Should Treat People Equitably Not Exacerbate Long-Standing Dispar-itiesrdquo Center for Democracy amp Technology httpscdtorginsightsdata-use-in-the-fight-against-covid-19-should-treat-people-equitably-not-exacerbate-long-standing-disparities
Greenwood Dazza Gregory Nadeau Pagona Tsormpatzoudi Bryan Wilson Jeffrey Saviano and Alex ldquoSandyrdquo Pentland 30 April 2020 ldquoCOVID-19 Contact Tracing Privacy Principlesrdquo MIT Computational Law Report Available at httpslawmitedupubcovid19contacttracingprivacy- principlesrelease7
Editorial Board 29 April 2020 ldquoShow Evidence That Apps for COVID-19 Contact-Tracing Are Secure and Effectiverdquo Nature 580 (7805) 563ndash563 httpsdoiorg101038d41586-020-01264-1
ldquoJoint Statement on Contact Tracingrdquo 19 April 2020 Available at httpscryptobriefingcomwp-contentuploads202004Joint-State-ment-from-Researcherspdf
Hinch Robert Will Probert Anel Nurtay Michelle Kendall Chris Wymant
108 Resources
Matthew Hall Katrina Lythgoe et al 16 April 2020 ldquoEffective Configura-tions of a Digital Contact Tracing App A Report to NHSXrdquo Available at httpsgithubcomBDI-pathogenscovid-19_instant_tracingblobmasterReportndashEffective Configurations of a Digital Contact Tracing Apppdf
Kahn Gilmor Daniel 16 April 2020 ldquoPrinciples for Technology-Assisted Contact-Tracingrdquo White Paper American Civil Liberties Union httpswwwacluorgreportaclu-white-paper-principles-technology-assisted- contact-tracing
Carroll Anna and Samantha Stroman 16 April 2020 ldquoFind My Friends in a Pandemic The Future of Contact Tracing in Americardquo CSIS Commis-sion on Strengthening Americarsquos Health Security April 16 2020 httpshealthsecuritycsisorgarticlesfind-my-friends-in-a-pandemic-the-future- of-contact-tracing-in-america
Electronic Privacy Information Center Testimony to Congress 15 April 2020 ldquoEPIC to Congress Establish Privacy Safeguards for Digital Contact Tracingrdquo httpsepicorgtestimonycongressEPIC-HEC-Contact-Tracing-Apr2020pdf
Crocker Andrew Kurt Opsahl and Bennett Cyphers 10 April 2020 ldquoThe Chal-lenge of Proximity Apps for COVID-19 Contact Tracingrdquo Electronic Fron-tier Foundation April 10 2020 httpswwwefforgdeeplinks202004challenge-proximity-apps-covid-19-contact-tracing
Hart Vi Divya Siddarth Bethan Cantrell Lila Tretikov Peter Eckersley John Langford Scott Leibrand et al 3 April 2020 ldquoOutpacing the Virus Digital Response to Containing the Spread of COVID-19 While Mitigat-ing Privacy Risksrdquo Whitepaper 5 COVID-19 Rapid Response Impact Initiative Edmond J Safra Center for Ethics httpsdrivegooglecomfiled1vIN2AX-DDNW-S0aHq8xs0RJ2jkR_CckXview
Raskar Ramesh Isabel Schunemann Rachel Barbar Kristen Vilcans Jim Gray Praneeth Vepakomma Suraj Kapa Andrea Nuzzo Rajiv Gupta et al 19 March 2020 ldquoApps Gone Rogue Maintaining Personal Privacy in an Epidemicrdquo White Paper Private Kit MIT httpsarxivorgpdf2003 08567pdf
General (not COVID-19 specific) Statements of Principles Legal Frameworks Other Information
Nuffield Council on Bioethics 2020 ldquoGuide to the Ethics of Surveillance and Quarantine for Novel Coronavirusrdquo httpswwwnuffieldbioethicsorgassetspdfsGuide-to-the-ethics-of-surveillance-and-quarantine-for-novel-coronaviruspdf
Resources 109
Schwartz Adam 2020 ldquoHow EFF Evaluates Government Demands for New Surveillance Powersrdquo Electronic Frontier Foundation April 3 2020 httpswwwefforgdeeplinks202004how-eff-evaluates-government-demands- new-surveillance-powers
WHO 2 April 2020 ldquoCoronavirus Disease 2019 (COVID-19)rdquo Situation Report 73 World Health Organization httpswwwwhointdocsdefault-sourcecoronavirusesituation-reports20200402-sitrep-73-covid-19pdf
Nuffield Council on Bioethics 17 March 2020 ldquoEthical Considerations in Responding to the COVID-19 Pandemicrdquo Rapid Policy Briefing Nuffield Council on Bioethics httpswwwnuffieldbioethicsorgassetspdfs Ethical-considerations-in-responding-to-the-COVID-19-pandemicpdf
Nuffield Council on Bioethics 28 January 2020 ldquoResearch in Global Health Emergencies Ethical Issuesrdquo httpswwwnuffieldbioethicsorgassetspdfsRGHE_full_report1pdf
Vota Wayan 4 December 2019 ldquoEvery African Countryrsquos National eHealth Strategy or Digital Health Policyrdquo ICT Works httpswwwictworksorgafrican-national-ehealth-strategy-policy
ENISA 3 December 2019 ldquoPseudonymisation Techniques and Best Practicesrdquo European Union Agency for Cybersecurity httpswwwenisaeuropaeupublicationspseudonymisation-techniques-and-best-practices
ldquoMobile Cellular Subscriptions (per 100 People)rdquo 2018 World Telecommuni-cationICT Development Report International Telecommunication Union httpsdataworldbankorgindicatorITCELSETSP2
Levinson-Waldman Rachel 2018 ldquoCellphones Law Enforcement and the Right to Privacyrdquo Brennan Center for Justice New York University School of Law httpswwwbrennancenterorgsitesdefaultfiles2019-08Report_Cell_Surveillance_Privacypdf
Code of Federal Regulations Part 46ndashProtection of Human Subjects (45 CFR 46 US Department of Health and Human Services)
General Data Protection Regulation (REGULATION (EU) 2016679 (GDPR) European Union)
European Data Protection Supervisor ldquoNecessity and Proportionalityrdquo httpsedpseuropaeudata-protectionour-worksubjectsnecessity- proportionality_en
ISOIEC 38505-12017 Information technologymdashGovernance of ITmdashGover-nance of datamdashPart 1 Application of ISOIEC 38500 to the governance of data (2017 International Organization for Standardization (ISO))
WHO 2016 ldquoWHO Guidance for Managing Ethical Issues in Infectious Dis-ease Outbreaksrdquo World Health Organization httpswwwwhointethicspublicationsinfectious-disease-outbreaksen
110 Resources
CIOMS 2016 ldquoInternational Ethical Guidelines for Health-Related Research Involving HumansndashGuideline 22 Use of Data Obtained from the Online Environment and Digital Tools in Health Related Researchrdquo Council for International Organizations of Medical Sciences httpsciomschwp- contentuploads201701WEB-CIOMS-EthicalGuidelinespdf
Principles for Digital Development 2016 ldquoPrinciplesrdquo httpsdigitalprinciplesorgprinciples
GA4GH 2014 ldquoFramework for Responsible Sharing of Genomic and Health-Related Datardquo Global Alliance for Genomics amp Health 9 Decem-ber 2014 httpswwwga4ghorggenomic-data-toolkitregulatory-ethics- toolkitframework-for-responsible-sharing-of-genomic-and-health-related- data
WHO and PATH 2013 ldquoPlanning an Information Systems Project A Toolkit for Public Health Managersrdquo World Health Organization amp PATH httpspathazureedgenetmediadocumentsTS_opt_ict_toolkitpdf
ASTHO 2012 ldquoPublic Health Collection Use Sharing and Protection of Informationrdquo Issue Brief ASTHO Legal Preparedness Series Association of State and Territorial Health Officials httpswwwasthoorgProgramsPreparednessPublic-Health-Emergency-LawPublic-Health-and-Informat- ion-Sharing-ToolkitCollection-Use-Sharing-and-Protection-Issue-Brief
Cavoukian Ann 2010 ldquoPrivacy by Design The 7 Foundational Principlesrdquo Ontario Canada Information and Privacy Commissioner of Canada httpsiapporgmediapdfresource_centerpbd_implement_7found_ principlespdf
PHLS 2002 ldquoPrinciples of the Ethical Practice of Public Healthrdquo Public Health Leadership Society httpswwwaphaorg-mediafilespdfmembergroupsethicsethics_brochureashx
National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research 1979 ldquoThe Belmont Report Ethical Principles and Guidelines for the Protection of Human Subjects of Researchrdquo US Dept of Health amp Human Services Office of Human Research Protections httpswwwhhsgovohrpregulations-and-policybelmont-reportread-the-bel mont-reportindexhtml
Reopening ProposalsPlans (Broad)
United Kingdom Cabinet Office 2020 ldquoOur Plan to Rebuild The UK Govern-mentrsquos COVID-19 Recovery Strategyrdquo 12 May 2020 httpswwwgovukgovernmentpublicationsour-plan-to-rebuild-the-uk-governments-covid- 19-recovery-strategyour-plan-to-rebuild-the-uk-governments-covid-19- recovery-strategyfourteen-supporting-programmes
Resources 111
Shannon Joel Lorenzo Reyes and Doyle Rice 2020 ldquoAre Lockdowns Being Relaxed in My State Herersquos How America Is Reopening amid the Coronavirus Pandemicrdquo USA TODAY May 21 2020 httpswww usatodaycomstorynewshealth20200419coronavirus-lockdown- reopening-states-us-texas-florida5155269002
Romer Paul 2020 ldquoRoadmap to Responsibly Reopen Americardquo 23 April 2020 httpsroadmappaulromernetpaulromer-roadmap-reportpdf
Governor Larry Hogan 24 April 2020 ldquoMaryland Strong Roadmap to Recoveryrdquo httpsgovernormarylandgovwp-contentuploads202004MD_Strongpdf
NGA and ASTHO 21 April 2020 ldquoRoadmap to Recovery A Public Health Guide for Governorsrdquo National Governors Association and American State and Territorial Health Officials httpswwwngaorgwp-contentuploads202004NGA-Reportpdf
Allen Danielle Julius Krein Ganesh Sitaraman and E Glen Weyl 2020 ldquoNational Covid-19 Testing Action Planrdquo The Rockefeller Foundation httpswwwrockefellerfoundationorgwp-contentuploads202004TheRockefellerFoundation_WhitePaper_Covid19_4_22_2020pdf
Edmond J Safra Center for Ethics Harvard UniverityndashWhite PapersAllen Danielle Sharon Block Joshua Cohen et al 20 April 2020 ldquoRoad-
map to Pandemic Resiliencerdquo httpsethicsharvardedufilescenter- for-ethicsfilesroadmaptopandemicresilience_updated_42020_0pdf
Allen Danielle Lucas Stanczyk Rajiv Sethi Glen Weyl 25 March 2020 ldquoWhen Can We Go Outrdquo httpsdrivegooglecomfiled1gf21eYeNWwrR9OO5nzxn1jlv-RTmHkt0view
Mulheirn Ian Sam Alvis Lizzie Insall James Browne Christina Palmou 20 April 2020 ldquoA Sustainable Exit Strategy Managing Uncertainty Minimis-ing Harmrdquo Tony Blair Institute for Global Change httpsinstituteglobalsitesdefaultfilesinline-filesA20Sustainable20Exit20Strategy2C 20Managing20Uncertainty20Minimising20Harmpdf
Emanuel Zeke Neera Tanden Adam Conner Erin Simpson Nicole Rap-fogel and Maura Calsyn 2020 ldquoA National and State Plan to End the Coronavirus Crisisrdquo Center for American Progress April 3 2020 httpswwwamericanprogressorgissueshealthcarenews20200403482613national-state-plan-end-coronavirus-crisis
Gottlieb Scott Caitlin Rivers Mark McClellan Lauren Silvis and Crystal Watson 2020 ldquoNational Coronavirus Response A Road Map to Reopen-ingrdquo American Enterprise Institute httpswwwaeiorgresearch-productsreportnational-coronavirus-response-a-road-map-to-reopening
112 Resources
Digital Contact Tracing Experiences from Other Countries
Multiple
Woodhams Samuel 12 May 2020 ldquoCOVID-19 Digital Rights Trackerrdquo Top10VPN httpswwwtop10vpncomresearchinvestigationscovid-19- digital-rights-tracker
Fahim Kareem Min Joo Kim and Steve Hendrix 2 May 2020 ldquoCellphone Monitoring Is Spreading with the Coronavirus So Is an Uneasy Tolerance of Surveillancerdquo Washington Post (Washington DC) httpswww washingtonpostcomworldcellphone-monitoring-is-spreading-with-the- coronavirus-so-is-an-uneasy-tolerance-of-surveillance20200502 56f14466-7b55-11ea-a311-adb1344719a9_storyhtml
Ikram Umar Christer Mjaringset MD Anne-Marie Boxall Mylaine Breton Ines Gravey Holly Krelle Veacuteronique Raimond and Reginald D Williams II 30 April 2020 ldquoWhat Can the US Learn from Innovative Strategies Used in Other Countries to Respond to COVID-19rdquo The Commonwealth Fund httpswwwcommonwealthfundorgblog2020what-can-us-learn-innovative-strategies-used-other-countries-respond-covid-19
Jens-Henrik Jeppesen and Pasquale Esposito 29 April 2020 ldquoCOVID-19 European Data Collection and Contact Tracing Measuresrdquo Center for Democracy amp Technology httpscdtorginsightscovid-19-european- data-collection-and-contact-tracing-measures
Africa CDC 9 April 2020 ldquoGuidance on Contact Tracing for COVID-19 Pan-demicrdquo Manuals Guidelines amp Frameworks African Union Africa CDC httpsafricacdcorgdownloadguidance-on-contact-tracing-for-covid-19- pandemic
Heneghan Carl Jon Brassey and Tom Jefferson 6 April 2020 ldquoCOVID-19 What Proportion Are Asymptomaticrdquo Centre for Evidence-Based Medi-cine httpswwwcebmnetcovid-19covid-19-what-proportion-are- asymptomatic
Kharpal Arjun 30 March 2020 ldquoUse of Surveillance to Fight Coronavirus Raises Concerns about Government Power after Pandemic Endsrdquo CNBC httpswwwcnbccom20200327coronavirus-surveillance-used-by- governments-to-fight-pandemic-privacy-concernshtml
Canada
ldquoCommissioner Publishes Framework to Assess Privacy-Impactful Initiatives in Response to COVID19rdquo Office of the Privacy Commissioner of Canada 17 April 2020 httpswwwprivgccaenopc-newsnews-and-announcements 2020an_200417
Resources 113
China
Kraemer Moritz U G Chia-Hung Yang Bernardo Gutierrez Chieh-Hsi Wu Brennan Klein David M Pigott Open COVID-19 Data Working Group Louis du Plessis Nuno R Faria Ruoran Li William P Hanage John S Brownstein Maylis Layan Alessandro Vespignani Huaiyu Tian Chris-topher Dye Oliver G Pybus Samuel V Scarpino ldquoThe Effect of Human Mobility and Control Measures on the COVID-19 Epidemic in Chinardquo Science 368(6490) 493-497 DOI 101126scienceabb4218
Bi Qifang Yongsheng Wu Shujiang Mei Chenfei Ye Xuan Zou Zhen Zhang Xiaojian Liu Lan Wei Shaun A Truelove Tong Zhang Wei Gao Cong Cheng Xiujuan Tang Xiaoliang Wu Yu Wu Binbin Sun Suli Huang Yu Sun Juncen Zhang Ting Ma Justin Lessler and Teijian Feng ldquoEpidemiol-ogy and Transmission of COVID-19 in 391 Cases and 1286 of Their Close Contacts in Shenzhen China A Retrospective Cohort Studyrdquo The Lancet April 27 2020 DOIhttpsdoiorg101016S1473-3099(20)30287-5
Sun Kaiyuan and Ceacutecile Viboud ldquoImpact of contact tracing on SARS-CoV-2 transmissionrdquo The Lancet April 27 2020 DOIhttpsdoiorg101016S1473-3099(20)30357-1
Ferretti Luca Chris Wymant Michelle Kendall Lele Zhao Anel Nurtay Lucie Abeler-Doumlrner Michael Parker David Bonsall and Christophe Fraser 2020 ldquoQuantifying SARS-CoV-2 Transmission Suggests Epidemic Control with Digital Contact Tracingrdquo Science 368 (6491) httpsdoiorg101126scienceabb6936
Gan Nectar and David Culver 2020 ldquoChina Is Fighting the Coronavirus with a Digital QR code Herersquos How It Worksrdquo CNN Business April 16 2020 httpswwwcnncom20200415asiachina-coronavirus-qr-code-intl-hnkindexhtml
Mozur Paul Raymond Zhong and Aaron Krolik 2020 ldquoIn Coronavirus Fight China Gives Citizens a Color Code with Red Flagsrdquo New York Times (New York NY) March 1 2020 httpswwwnytimescom20200301businesschina-coronavirus-surveillancehtml
Germany
Schwartz Matthew S 2020 ldquoGermany Backs Away from Compiling Coro-navirus Contacts in a Central Databaserdquo NPR April 27 2020 https wwwnprorgsectionscoronavirus-live-updates20200427846046185 germany-backs-away-from-compiling-coronavirus-contacts-in-a-central- databaseutm_medium=RSSamputm_campaign=news
Busvine Douglas and Andreas Rinke 2020 ldquoGermany Flips to Apple-Google
114 Resources
Approach on Smartphone Contact Tracingrdquo Reuters April 26 2020 httpswwwreuterscomarticleus-health-coronavirus-europe-techgermany-flips-on-smartphone-contact-tracing-backs-apple-and-google-idUSKCN22807J
Hong Kong
ldquo lsquoStayHomeSafersquo Mobile App User Guiderdquo The Government of the Hong Kong
Special Administrative Region May 20 2020httpswwwcoronavirusgov hkengstay-home-safehtml
India
Dixit Pranav 2020 ldquoIndiarsquos Contact Tracing App Is All But Mandatory So This Programmer Hacked It So That He Always Appears Saferdquo BuzzFeed News May 12 2020 httpswwwbuzzfeednewscomarticlepranavdixitindia-aarogya-setu-hacked
Greenberg Andy 2020 ldquoIndiarsquos Covid-19 Contact Tracing App Could Leak Patient Locationsrdquo WIRED May 6 2020 httpswwwwiredcomstoryindia-covid-19-contract-tracing-app-patient-location-privacy
Alderson Elliot ldquoAarogya Setu The Story of a Failurerdquo Medium May 6 2020 httpsmediumcomfs0c131yaarogya-setu-the-story-of-a-failure- 3a190a18e34
OrsquoNeill Patrick Howell ldquoIndia Is Forcing People to Use Its COVID App Unlike Any Other Democracyrdquo MIT Technology Review May 6 2020 httpswwwtechnologyreviewcom202005071001360india-aarogya- setu-covid-app-mandatory
Israel
Hendrix Steve and Ruth Eglash 2020 ldquoIsrael Is Using Cellphone Surveil-lance to Warn Citizens You May Already Be Infectedrdquo Washington Post (Washington DC) March 19 2020 httpswwwwashingtonpostcomworldmiddle_eastisrael-is-using-cellphone-surveillance-to-warn-citizens-you-may-already-be-infected2020031968267294-69e7-11ea-b199-3a9799c54512_storyhtml
Singapore
ldquoTraceTogether Safer Togetherrdquo A Singapore Government Agency Website accessed May 21 2020 httpswwwtracetogethergovsg
Resources 115
South Korea
Korean Ministry of Health and Welfare ldquoConfirmed Patient Movement Path Websiterdquo Central Accident Remediation Headquarters accessed May 21 2020 httpncovmohwgokrbdBoardList_RealdobrdId=1ampbrdGu-bun=12ampncvContSeq=ampcontSeq=ampboard_id=ampgubun=
Kim Max S 2020 ldquoSeoulrsquos Radical Experiment in Digital Contact Tracingrdquo The New Yorker April 17 2020 httpswwwnewyorkercomnewsnews-deskseouls-radical-experiment-in-digital-contact-tracing
Watson Ivan and Sophie Jeong 2020 ldquoCoronavirus Mobile Apps Are Surging in Popularity in South Koreardquo CNN Business February 28 2020 httpseditioncnncom20200228techkorea-coronavirus-tracking-appsindexhtml
United Kingdom
Hern Alex and Kate Proctor 2020 ldquoUK May Ditch NHS Contact-Tracing App for Apple and Google Modelrdquo The Guardian May 7 2020 httpswwwtheguardiancomtechnology2020may07uk-may-ditch-nhs-contact- tracing-app-for-apple-and-google-model
Lovejoy Ben 2020 ldquoHands-on with UKrsquos NHS Contact Tracing App as the Test Goes Liverdquo 9-5 Mac Blog May 7 2020 https9to5maccom2020 0507nhs-contact-tracing
ldquoCoronavirus Test Track and Trace Plan Launched on Isle of Wightrdquo Depart-ment of Health and Social Care Press Release May 4 2020 httpswww govukgovernmentnewscoronavirus-test-track-and-trace-plan-launched- on-isle-of-wight
Leprince-Ringuet Daphne 2020 ldquoContact-Tracing Apps Why the NHS Said No to Apple and Googlersquos Planrdquo ZDNet April 28 2020 httpswww zdnetcomarticlecontact-tracing-apps-why-the-nhs-said-no-to-apple- and-googles-plan
Specific Digital ProductsApps
COVID-19
Vota Wayan 2020 Additional Proposed Coronavirus Solutions Google Doc-ument httpsdocsgooglecomspreadsheetsd15hkhdtGNzx7oHkO8Y 2MOiY83JsHjqxL4MhMGvlA_J6Ieditgid=579623365
Apple and Google nd ldquoPrivacy-Preserving Contact Tracingrdquo Apple Accessed May 19 2020 httpswwwapplecomcovid19contacttracing
116 Resources
Woodhams Samuel 12 May 2020 ldquoCOVID-19 Digital Rights Trackerrdquo Top10VPN httpswwwtop10vpncomresearchinvestigationscovid-19-digital-rights-tracker
Starobinski David and Johannes Becker 30 April 2020 ldquoHow Apple and Goo-gle Will Let Your Phone Warn You If Yoursquove Been Exposed to the Corona-virusrdquo The Conversation Accessed May 21 2020 httptheconversation comhow-apple-and-google-will-let-your-phone-warn-you-if-youve-been-exposed-to-the-coronavirus-136597
Morrison Sara 2020 ldquoThe Apple-Google Contact Tracing Tool Gets a Beta Release and a New Risk Level Featurerdquo Vox April 24 2020 httpswwwvoxcomrecode202042421234420apple-google-contact-tracing- exposure-notification-update
Commonwealth Centre for Digital Health 9 April 2020 ldquo[Webinar] CWCDH Digital Response to COVID-19rdquo 5040 httpswwwyoutubecomwatchv=ZyE_KRWLtC8ampfeature=youtube
Johns Hopkins Medicine 23 April 2020 ldquoJohns Hopkins Medicine Remote Monitoring Program for Faculty Staff and Providers Exposed to COVID-19rdquo Johns Hopkins Medicine Occupational Health Services httpswww hopkinsmedicineorghsecovid19_emocha
MIT nd ldquoPrivate Kit Safe Paths Privacy-by-Design Covid19 Solutions Using GPS+Bluetooth for Citizens and Public Health Officialsrdquo Safepaths Accessed May 20 2020 httpssafepathsmitedu
Related Disease Detection Projects
ldquoCOVID Controlrdquo Google Play accessed May 21 2020 httpsplaygoogle comstoreappsdetailsid=jhueduJohnsHopkinsCOVIDControl
ldquoCOVID Symptom Trackerrdquo Created by Massachusetts General Hospital the Harvard TH Chan School of Public Health Kingrsquos College London and Stanford University School of Medicine accessed May 21 2020 httpscovidjoinzoecomus
ldquoSee How Your Community Is Moving around Differently Due to COVID-19rdquo Google COVID-19 Mobility Reports accessed May 21 2020 httpswww googlecomcovid19mobility
Drew David A Long H Nguyen Claire J Steves Cristina Menni Maxim Freydin Thomas Varsavsky Carole H Sudre M Jorge Cardoso Sebastien Ourselin Jonathan Wolf Tim D Spector Andrew T Chan and COPE Consortium 2020 ldquoRapid Implementation of Mobile Technology for Real-Time Epidemiology of COVID-19rdquo Science 05 May 2020 DOI 101126scienceabc0473
Resources 117
Tress Luke 2020 ldquoMaccabi Medial EarlySign Develop Algorithm to Identify High-Risk COVID-19 Casesrdquo Times of Israel April 22 2020 httpswwwtimesofisraelcommaccabi-medial-earlysign-develop-algo rithm-to-identify-high-risk-covid-19-cases
ldquoPCR Diagnostic Testing for SARS-CoV-2rdquo Center for Health Security Johns Hopkins Bloomberg School of Public Health last modified April 17 2020 httpswwwcenterforhealthsecurityorgresourcesCOVID-19COVID-19-fact-sheets200130-nCoV-diagnostics-factsheetpdf
Radin Jennifer M Nathan E Wineinger Eric J Topol and Steven R Steinhubl 2020 ldquoHarnessing Wearable Device Data to Improve State-Level Real-Time Surveillance of Influenza-Like Illness in the USA A Population-Based Studyrdquo The Lancet January 16 2020 DOIhttpsdoiorg101016S2589-7500(19)30222-5
Polling
Hargittai Eszter Minh Hao Nguyen Jaelle Fuchs Jonathan Gruber Will Marler Amanda Hunsaker and Goumlkccedile Karaoglu 2020 ldquoCovid-19 Study on Digital Media and the Coronavirus Pandemicrdquo Internet Use and Society Division Institute of Communication and Media Research University of Zurich httpwebuseorgcovid
Navigator Research 2020 ldquoPublic Opinion on Coronavirus Navigator Updaterdquo Navigating Coronavirus (blog) May 21 2020 https navigatorresearchorgnavigating-coronavirus
Russonello Giovanni 2020 ldquoBig Government For Now Most Americans Say Bring It Onrdquo The New York Times May 1 2020 sec US httpswww nytimescom20200501uspoliticscoronavirus-spending-pollshtml
ldquoWashington Post-University of Maryland National Pollrdquo 2020 Washington Post April 21 2020 httpswwwwashingtonpostcomcontext washington-post-university-of-maryland-national-poll-april-21-26-20203583b4e9-66be-4ed6-a457-f6630a550ddf
Kirzinger Ashley Liz Hamel Cailey Muntildeana Audrey Kearney and Mollyann Brodie 2020 ldquoKFF Health Tracking Poll ndash Late April 2020 Coronavirus Social Distancing and Contact Tracingrdquo Kaiser Family Foundation (blog) April 24 2020 httpswwwkfforgcoronavirus-covid-19issue-briefkff-health-tracking-poll-late-april-2020
NYC DOH 2020 ldquoAge-Adjusted Rates of Lab Confirmed COVID-19 Non-hospitalized Cases Estimated Non-Fatal Hospitalized Cases and Persons Known to Have Died per 100000 by RaceEthnicity Grouprdquo New York
118 Resources
City Department of Health httpswww1nycgovassetsdohdownloadspdfimmcovid-19-deaths-race-ethnicity-04242020-1pdf
Anderson Monica and Brooke Auxier 2020 ldquoMost Americans Donrsquot Think Cellphone Tracking Will Help Limit COVID-19 Are Divided on Whether Itrsquos Acceptablerdquo Pew Research Center (blog) April 16 2020 httpswwwpewresearchorgfact-tank20200416most-americans-dont-think-cellphone-tracking-will-help-limit-covid-19-are-divided-on-whether-its-acceptable
Daly Kyle 2020 ldquoExclusive Americans Wary of Giving up Data to Fight Coronavirusrdquo Axios April 3 2020 httpswwwaxioscomexclusive- americans-wary-of-giving-up-data-to-fight-coronavirus-330fc1d9-8b99-4a51-871b-25ee0e0591f2html
Elliott Douglas J Ana Kreacic Lorenzo Milans del Bosch and Lisa Quest nd ldquoData-Sharing in the Time of Coronavirusrdquo Oliver Wyman Forum Accessed May 19 2020 httpswwwoliverwymanforumcomfuture-of-data2020aprdata-sharing-in-the-time-of-coronavirushtml
Auxier Brooke Lee Rainie Monica Anderson Andrew Perrin Madhu Kumar and Erica Turner 2019 ldquoAmericans and Privacy Concerned Confused and Feeling Lack of Control Over Their Personal Informationrdquo Pew Research
Center Internet Science amp Tech (blog) November 15 2019 httpswww pewresearchorginternet20191115americans-and-privacy-concerned- confused-and-feeling-lack-of-control-over-their-personal-information
EPIC 2020 ldquoPublic Opinion on Privacyrdquo Electronic Privacy Information Center January 22 2020 httpsepicorgprivacysurvey
Pew Research Center 2020 ldquoDemographics of Mobile Device Ownership and Adoption in the United Statesrdquo Internet amp Technology Pew Research Cen-ter httpswwwpewresearchorginternetfact-sheetmobile
Rodrigues Rafaela Alina Husain Amanda Couture-Carron Leslye E Orloff and Nawal H Ammar 2018 ldquoPromoting Access to Justice for Immigrant and Limited English Proficient Crime Victims in an Age of Increased Immi-gration Enforcement Initial Report from a 2017 National Surveyrdquo Wash-ington DC National Immigrant Womenrsquos Advocacy Project American University Washington College of Law httpniwaplibrarywclamericaneduwp-contentuploadsImmigrant-Access-to-Justice-National-Reportpdf
Center for Survey Measurement (CSM) MEMORANDUM FOR Associate Directorate for Research and Methodology (ADRM) 2017 ldquoRespondent Confidentiality Concernsrdquo September 20 2017 httpswww2censusgovcacnacmeetings2017-11Memo-Regarding-Respondent-Confidentiality- Concernspdf
Resources 119
Pew Research Center 2017 ldquoUS Muslims Concerned About Their Place in Society but Continue to Believe in the American Dreamrdquo Religion amp Public Life Pew Research Center httpswwwpewforumorg20170726findings-from-pew-research-centers-2017-survey-of-us-muslims
Popular Press
Dixit Pranav 2020 ldquoIndiarsquos Contact Tracing App Is All But Mandatory So This Programmer Hacked It So That He Always Appears Saferdquo BuzzFeed News May 12 2020
Mills Rodrigo Chris 2020 ldquoDigital Contact Tracing Is Becoming Available but Is It Effectiverdquo The Hill May 7 2020 httpsthehillcompolicytechnology 496498-digital-contact-tracing-is-becoming-available-but-is-it-effective
Ingram David 2020 ldquoApple Google Push Makers of Coronavirus Apps Not to Record User Locationrdquo May 4 2020 NBC News Accessed May 19 2020 httpswwwnbcnewscomtechtech-newscoronavirus-apps-won-t-be-able- record-users-location-apple-n1199586
McMinn Sean 2020 ldquoMobile Phone Data Show More Americans Are Leav- ing Their Homes Despite Ordersrdquo NPR May 1 2020 httpswwwnpr org20200501849161820mobile-phone-data-show-more-americans- are-leaving-their-homes-despite-orders
Valentino-DeVries Jennifer Natasha Singer and Aaron Krolik 2020 ldquoA Scram-ble for Virus Apps That Do No Harmrdquo The New York Times April 29 2020 sec Business httpswwwnytimescom20200429 businesscoronavirus-cellphone-apps-contact-tracinghtml
Giglio Mike 2020 ldquoWould You Sacrifice Your Privacy to Get Out of Quaran-tinerdquo The Atlantic April 22 2020 httpswwwtheatlanticcompoliticsarchive202004coronavirus-pandemic-privacy-civil-liberties-911609172
Bradshaw Tim 2020 ldquo2 Billion Phones Cannot Use Google and Apple Con-tact-Tracing Tech | Ars Technicardquo Ars Technica April 20 2020 httpsarstechnicacomtech-policy2020042-billion-phones-cannot-use-google-and-apple-contract-tracing-tech
Reston Maeve Kristina Sgueglia and Cheri Mossburg 2020 ldquoGovernors on East and West Coasts Form Pacts to Decide When to Reopen Economiesrdquo CNN Politics April 13 2020 httpswwwcnncom20200413politicsstates-band-together-reopening-plansindexhtml
Cochrane Emily Claire Cain Miller and Jim Tankersley 2020 ldquoTrump Admin-istration Scales Back Paid Leave in Coronavirus Relief Lawrdquo The New York Times April 2 2020 sec US httpswwwnytimescom20200402uspoliticscoronavirus-paid-leavehtml
120 Resources
Kim Nemo 2020 ldquolsquoMore Scary than Coronavirusrsquo South Korearsquos Health Alerts Expose Private Livesrdquo The Guardian March 6 2020 sec World news httpswwwtheguardiancomworld2020mar06more-scary- than-coronavirus-south-koreas-health-alerts-expose-private-lives
Hickey Matt 2014 ldquoCarriers Can Now Install Apps On Android Handsets Without Customersrsquo Permissionrdquo Forbes December 1 2014 httpswww forbescomsitesmatthickey20141201carriers-can-now-install-apps-on- android-handsets-without-customers-permission
Commentaries
Giubilini Alberto 2020 ldquoContact-Tracing Apps and the Future COVID-19 Vaccination Should Be Compulsory Social Technological and Pharmaco-logical Immunisationrdquo Practical Ethics (blog) May 6 2020 httpblo gpracticalethicsoxacuk202005contact-tracing-apps-and-the-future- covid-19-vaccination-should-be-compulsory-social-technological-and- pharmacological-immunisation
Landau Susan Christy Lopez and Laura Moy 2020 ldquoThe Importance of Equity in Contact Tracingrdquo Lawfare (blog) May 1 2020 httpswww lawfareblogcomimportance-equity-contact-tracing
Schaefer G Owen and Angela Ballantyne 2020 ldquoDownloading COVID-19 Contact Tracing Apps Is a Moral Obligationrdquo Journal of Medical Ethics
Blog (blog) May 4 2020 httpsblogsbmjcommedical-ethics20200504downloading-covid-19-contact-tracing-apps-is-a-moral-obligation
OrsquoNeill Patrick Howell 2020 ldquoFive Things We Need to Do to Make Contact Tracing Really Workrdquo MIT Technology Review April 28 2020 httpswwwtechnologyreviewcom202004281000714five-things-to-make- contact-tracing-work-covid-pandemic-apple-google
Doffman Zak 2020 ldquoCOVID-19 Contact Tracing Why Apple And Google Canrsquot Make This Workrdquo Forbes April 27 2020 httpswwwforbescomsiteszakdoffman20200427this-is-the-contact-tracing-worry-even-apple- and-google-cant-resolve
All Tech is Human 2020 The Ethics of Contact Tracing for COVID-19 httpswwwyoutubecomwatchv=59mKUAVDhdkampt=626s
Canca Cansu 2020 ldquoWhy lsquoMandatory Privacy-Preserving Digital Contact Tracingrsquo Is the Ethical Measure against COVID-19rdquo Medium May 4 2020 httpsmediumcomcansucancawhy-mandatory-privacy-preserv-ing-digital-contact-tracing-is-the-ethical-measure-against-covid-19-a0d-143b7c3b6
Goodman Bryce 2020 ldquoCOVID and Contact Tracing When Social Justice
Resources 121
Demands Mass Surveillancerdquo Medium April 10 2020 httpsmediumcombwgoodmancovid-and-contact-tracing-when-social-justice-demands-mass-surveillance-18d419b8cc5
Gray Rosie and Caroline Haskins 2020 ldquoThey Were Opposed To Govern-ment Surveillance Then The Coronavirus Pandemic Beganrdquo BuzzFeed News March 30 2020 httpswwwbuzzfeednewscomarticlerosiegraythey-were-opposed-to-government-surveillance-then-the
Cegłowski Maciej 2020 ldquoWe Need A Massive Surveillance Programrdquo Idle
Words (blog) March 23 2020 httpsidlewordscom202003we_need_ a_massive_surveillance_programhtm
Academic Literature
COVID-19 Specific
Abeler J Baumlcker M Buermeyer U Zillessen H (2020) COVID-19 Contact Tracing and Data Protection Can Go Together JMIR mHealth and uHealth 8(4) e19359 doi 10219619359
Altmann S Milsom L Zillessen H et al (2020) Acceptability of App-Based Contact Tracing for COVID-19 Cross-Country Survey Evidence Preprint
Berke A Bakker M Vepakomma P Larson K Pentland A (2020) Assessing Disease Exposure Risk with Location Data A Proposal for Cryptographic Preservation of Privacy arXiv arXiv200314412ndashMarch 2020
Bonsail D Parker M Fraser C (2020) Sustainable Containment of COVID-19 Using Smartphones in China Scientific and Ethical Underpinnings for Implementation of Similar Approaches in Other Settings Unpublished working paper
Bradshaw WJ Alley EC Huggins JH Lloyd AL Esvelt KM (2020) Bidirectional Contact Tracing Is Required for Reliable COVID-19 Control Preprint via MedRxiv
Braithwaite I Callender T Bullock M Aldridge R (2020) Automated and Semi-Automated Contact Tracing Protocol for a Rapid Review of Avail-able Evidence and Current Challenges to Inform the Control of COVID-19 Preprint via medRxiv doi httpsdoiorg1011012020041420063636
Bulchandani Bannerjee V Shivam S Moudgalya S Sondhi SL (2020) Digital Herd Immunity and COVID-19 Preprint via medRxiv doi httpsdoiorg1011012020041520066720
Cheng H Jian S Liu D (2020) Contact Tracing Assessment of COVID-19 Transmission Dynamics in Taiwan and Risk at Different Exposure Periods Before and After Symptom Onset JAMA Internal Medicine doi101001jamainternmed20202020
122 Resources
Cho H Ippolito D Yu YW (2020) Contact Tracing Mobile Apps for COVID-19 Privacy Considerations and Related Trade-offs httpsarxivorgpdf200311511pdf
Devakumar D Geordan S Bhopal SS Abubakar I (2020) Racism and dis-crimination in COVID-19 responses The Lancet 395(10231) 1194 doi 101016S0140-6736(20)30792-3
Drew D Nguyen L Steves C et al (2020) Rapid Implementation of Mobile Technology for Real-Time Epidemiology of COVID-19 Science published online May 5 2020
Ferretti L Wymant C Kendall M et al (2020) Quantifying SARS-CoV-2 Trans-mission Suggests Epidemic Control with Digital Contact Tracing Science doi 101126scienceabb6936
Fraser C Abeler-Doumlrner L Ferretti L et al (2020) Digital Contact Tracing Comparing the Capabilities of Centralised and Decentralised Data Archi-tectures to Effectively Suppress the COVID-19 Epidemic While Maximizing Freedom of Movement and Maintaining Privacy Preprint
Leith DJ Farrell S (2020) Coronavirus Contact Tracing Evaluating The Poten-tial Of Using Bluetooth Received Signal Strength For Proximity Detection Preprint
Jayant Limaye R Sauer M Ali J et al (2020) Building Trust While Influencing Online COVID-19 Content iIn the Social Media World The Lancet Digital Health
Mahmood S Hasan K Colder Carras M Labrique A (2020) Global Prepared-ness Aagainst COVID-19 We Must Leverage the Power of Digital Health JMIR Public Health Surveill 20206(2)e18980 DOI 10219618980
Mello M Wang CJ (2020) Ethics and Governance for Digital Disease Surveil-lance Science 11 May 2020 eabb9045 DOI 101126scienceabb9045
Park S Jeehyun Choi G Ko H (2020) Information TechnologyndashBased Tracing Strategy in Response to COVID-19 in South KoreamdashPrivacy Controversies JAMA doi101001jama20206602
Parker M Fraser C Abeler-Doumlrner L Bonsall D (2020) Ethics of Instantaneous Contract Tracing Using Mobile Phone Apps in the Control of the COVID-19 Pandemic Journal of Medical Ethics Published Online May 4 2020
Ethics and Digital Disease Detection
Aiello A Renson A Civich P (2020) Social Mediandash and Internet-Based Dis-ease Surveillance for Public Health Annual Review of Public Health 41 101ndash118 doi 101146annurev-publhealth-040119-094402
Ali J DiStefano M Coates McCall I et al (2019) Ethics of Mobile Phone Sur-
Resources 123
veys to Monitor Non-Communicable Disease Risk Factors in Low- and Middle-Income Countries A Global Stakeholder Survey Global Public Health 14(8) 1167ndash1181
Ali J Labrique A Gionfriddo K et al (2017) Ethics Considerations in Global Mobile Phone-Based Surveys of Noncommunicable Diseases A Concep-tual Exploration Journal of Medical Internet Research 19(5) e110 doi 102196jmir7326
Brownstein J Freifeld C Madoff L (2009) Digital Disease DetectionndashHar-nessing the Web for Public Health Surveillance New England Journal of Medicine 360(21) 2153ndash2157 doi 101056NEJMp0900702
Danquah LO Hasham N MacFarlane M et al (2019) Use of a Mobile Application for Ebola Contact Tracing and Monitoring in Northern Sierra Leone A Proof-of-Concept Study BMC Infectious Diseases 19 810
Degeling C Carter S van Oijen A et al (2020) Community Perspectives on the Benefits and Risks of Technologically Enhanced Communicable Disease Surveillance Systems A Report on Four Community Juries BMC Medical Ethics 21 31 doi 101186s12910-020-00474-6
DeJong B Badou G Luten J et al (2019) Ethical Considerations for Movement Mapping to Identify Disease Transmission Hotspots Emerging Infectious Diseases 25(7) e181421 doi 103201eid2507181421
Denecke K (2017) An Ethical Assessment Model for Digital Disease Detec-tion Technologies Life Sciences Society and Policy 13 16 doi 101186s40504-017-0062-x
Genevieve LD Martani A Wangmo T et al (2019) Participatory Disease Surveillance Systems Ethical Framework Journal of Medical Internet Research 21(5) e12273 doi10219612273
Gilbert G Degeling C Johnson J (2017) Communicable Disease Surveillance Ethics in the Age of Big Data and New Technology Asian Bioethics Review 11 173-187 doi 101007s41649-019-00087-1
Iwaya LH Li J Fischer-Hubner S et al (2019) E-Consent for Data Privacy Consent Management for Mobile Health Technologies in Public Health Surveys and Disease Surveillance Studies in health technology and infor-matics 264 1223-1227 doi 103233SHTI190421
Kostkova P (2018) Disease Surveillance Data Sharing for Public Health The Next Ethical Frontiers Life Sciences Society and Policy 14 16 doi 101186s40504-018-0078-x
Mahmood S Hasan K Colder Carras M Labrique A (2020) Global Prepared-ness against COVID-19 We Must Leverage the Power of Digital Health JMIR Public Health Surveillance 6(2)e18980
124 Resources
Smolinski MS Crawley AW Baltrusaitis K et al (2015) Flu Near You Crowd-sourced Symptom Reporting Spanning 2 Influenza Seasons AJPH 105(10) 2124ndash2130
Wojcik O Brownstein J Chunara R Johansson M (2014) Public Health for the People Participatory Infectious Disease Surveillance in the Digital Age Emerging Themes in Epidemiology 11 7 doi 1011861742-7622-11-7
Other Works
Bernstein J Holroyd TA Atwell JE et al (2019) Rockland Countyrsquos Proposed Ban against Unvaccinated Minors Balancing Disease Control Trust and Liberty Vaccine 37(30) 3933ndash3935
Berry SM Petzold EA Dull P et al (2017) A Response Adaptive Randomiza-tion Platform Trial for Efficient Evaluation of Ebola Virus Treatments A Model for Pandemic Response Clinical Trials 13(1) 22ndash30 doi 101177 1740774515621721
Beukenhorst AL Schultz DM McBeth J (2017) Using Smartphones for Research outside Clinical Settings How Operating Systems App Develop-ers and Users Determine Geolocation Data Quality in mHealth Studies MEDINFO 2017 Precision Healthcare through Informatics
Bourne P (2015) Confronting the Ethical Challenges of Big Data in Public Health Plos Computational Biology 11(2) e1004073 doi 101371journal pcbi1004073
Doerr M Suver C Wilbanks J (2016) Developing a Transparent Par ticipant-Navigated Electronic Informed Consent for Mobile-Mediated Research (April 22 2016) Available at SSRN httpsssrncomabstract=2769129 or httpdxdoiorg102139ssrn2769129
Dredze M Paul MJ Bergsma S Tran H (2013) Carmen A Twitter Geolocation System with Applications to Public Health Expanding the Boundaries of Health Informatics Using Artificial Intelligence Papers from the AAAI 2013 Workshop
Eckhoff PA Tatem AJ (2015) Digital methods in epidemiology can transform disease controlInternational Health Volume 7 Issue 2 March 2015 Pages 77ndash78 httpsdoiorg101093inthealthihv013
Edelstein M Lee L Herten-Crabb A Heymann D Harper D (2018) Strength-ening Global Public Health Surveillance through Data and Benefit Sharing Emerging Infectious Diseases 24(7) 1324ndash1330 doi 103201eid2407151830
Faden R Beauchamp T (1986) A History and Theory of Informed Consent Oxford University Press ISBN 9780199748655
Resources 125
Fairchild A Bayer R (2004) Ethics and the Conduct of Public Health Surveil-lance Science 303(5658) 631ndash632
Flanagan M Howe DC Nissenbaum H (2008) Embodying Values in Technol-ogy Theory and Practice In Information Technology and Moral Philoso-phy van den Hoven J amp Weckert J (eds) Cambridge Cambridge Univer-sity Press 322ndash353
Fraccaro P Beukenhorst A Sperrin M et al (2019) Digital Biomarkers from Geolocation Data in Bipolar Disorder and Schizophrenia A Systematic Review Journal of the American Medical Informatics Association Volume 26 Issue 11 November 2019 Pages 1412ndash1420
Furlanello C Merler S Menegon S et al (2002) New WEBGIS Technologies for Geo-location of Epidemiological Data An Application for the Sur-veillance of the Risk of Lyme borreliosis Disease In Giornale Italiano di Aritmologia e Cardiostimolazione Special Issue Proceedings of the 15th Int Congress ldquoThe New Frontiers of Arrhythmiasrdquo v 5 n1 Mar 2002 241ndash245
Gibson DG Wosu AC Pariyo GW et al (2019) Effect of Airtime Incentives on Response and Cooperation Rates in Non-communicable Disease Interactive Voice Response Surveys Randomised Controlled Trials in Bangladesh and Uganda BMJ Global Health 4(5)
Knobel C Bowker GC (2011) Computing Ethics Values in Design Communi-cations of the acm 54(7) 26ndash28
Lee R Cui RR Muessig KE et al (2015) Incentivizing HIVSTI Testing A Sys-tematic Review of the Literature AIDS and Behavior 18(5) 905ndash912 doi 101007s10461-013-0588-8
Lorchan LT Wyatt J (2014) mHealth and Mobile Medical Apps A Frame-work to Assess Risk and Promote Safer Use Journal of Medical Internet Research 16(9) e210 doi 102196jmir3133
Mathews S McShea M Hanley C et al (2019) Digital Health A Path to Vali-dation Digital Medicine 238 doi 101038s41746-019-0111-3
Moore S Tasse A Thorogood A et al (2017) Consent Processes for Mobile App Mediated Research Systematic Review JMIR mHealth amp uHealth 5(8) e126 DOI 102196mhealth7014
Pallman P Bedding AW Choodari-Oskooei B et al (2018) Adaptive Designs in Clinical Trials Why Use Them and How to Run and Report Them BMC Medicine 16(29)
Rennie S Buchbinder M Juengst E et al (2020) Scraping the Web for Public Health Gains Ethical Considerations from a lsquoBig Datarsquo Research Project on HIV and Incarceration Public Health Ethics doi 101093phephaa006
126 Resources
Rithalia A McDaid C Suekarran S (2009) Impact of Presumed Consent for Organ Donation on Donation Rates A Systematic Review BMJ 2009 338
Singer E Ye C (2012) The Use and Effects of Incentives in Surveys The ANNALS of the American Academy of Political and Social Science 645(1) 112ndash141
Vayena E Blassime A (2018) Health Research with Big Data Time for Systemic Oversight J Law Med Ethics 2018 Mar 46(1) 119ndash129
Vayena E Mastroianni A Kahn J (2012) Ethical Issues in Health Research with Novel Online Sources American Journal of Public Health 102(12) 2225ndash2230 doi 102105AJPH2012300813
Vayena E Salatheacute M Madoff L Brownstein J (2015) Ethical Challenges of Big Data in Public Health Plos Computational Biology 11(2) e1003904 doi 101371journalpcbi1003904
Xafis V Schaefer GO Labude MK et al (2019) An Ethics Framework for Big Data in Health and Research Asian Bioethics Review volume 11 pages 227ndash254(2019)
Works Cited
127
Africa CDC 2020 ldquoGuidance on Contact Tracing for COVID-19 Pandemicrdquo Manuals Guidelines amp Frameworks African Union Africa CDC httpsafricacdcorgdownloadguidance-on-contact-tracing-for-covid-19- pandemic
Aiello Allison E Audrey Renson and Paul N Zivich 2020 ldquoSocial Mediandash and Internet-Based Disease Surveillance for Public Healthrdquo Annual Review of Public Health 41 (1) 101ndash18 httpsdoiorg101146annurev-publhealth-040119-094402
Ali Joseph Michael J DiStefano Iris Coates McCall Dustin G Gibson Gulam Muhammed Al Kibria George W Pariyo Alain B Labrique and Adnan A Hyder 2019 ldquoEthics of Mobile Phone Surveys to Monitor Non-Communicable Disease Risk Factors in Low- and Middle-Income Countries A Global Stakeholder Surveyrdquo Global Public Health 14 (8) 1167ndash81 httpsdoiorg1010801744169220191566482
Ali Joseph Alain B Labrique Kara Gionfriddo George Pariyo Dustin G Gibson Bridget Pratt Molly Deutsch-Feldman and Adnan A Hyder 2017 ldquoEthics Considerations in Global Mobile Phone-Based Surveys of Noncommunicable Diseases A Conceptual Explorationrdquo Journal of Medi-cal Internet Research 19 (5) e110 httpsdoiorg102196jmir7326
Allen Danielle Julius Krein Ganesh Sitaraman and E Glen Weyl 2020 ldquoNational Covid-19 Testing Action Planrdquo The Rockefeller Foundation httpswwwrockefellerfoundationorgwp-contentuploads202004TheRockefellerFoundation_WhitePaper_Covid19_4_22_2020pdf
Altmann Samuel Luke Milsom Hannah Zillessen Raffaele Blasone Frederic Gerdon Ruben Bach Frauke Kreuter Daniele Nosenzo Severine Tous-saert and Johannes Abeler 2020 ldquoAcceptability of App-Based Contact Tracing for COVID-19 Cross-Country Survey Evidencerdquo MedRxiv May 2020050520091587 httpsdoiorg1011012020050520091587
Anderson Monica and Brooke Auxier 2020 ldquoMost Americans Donrsquot Think Cellphone Tracking Will Help Limit COVID-19 Are Divided on Whether
128 Works Cited
Itrsquos Acceptablerdquo Pew Research Center (blog) April 16 2020 https wwwpewresearchorgfact-tank20200416most-americans-dont-think-cellphone-tracking-will-help-limit-covid-19-are-divided-on-whether-its- acceptable
Anderson Monica and Andrew Perrin 2017 ldquoDisabled Americans Less Likely to Use Technologyrdquo Pew Research Center (blog) April 7 2017 httpswwwpewresearchorgfact-tank20170407disabled-americans- are-less-likely-to-use-technology
Apple and Google nd ldquoPrivacy-Preserving Contact Tracingrdquo Apple Accessed May 19 2020 httpswwwapplecomcovid19contacttracing
ASTHO 2012 ldquoPublic Health Collection Use Sharing and Protection of Informationrdquo Issue Brief ASTHO Legal Preparedness Series Association of State and Territorial Health Officials httpswwwasthoorgProgramsPreparednessPublic-Health-Emergency-LawPublic-Health-and- Information-Sharing-ToolkitCollection-Use-Sharing-and-Protection- Issue-Brief
Auxier Brooke Lee Rainie Monica Anderson Andrew Perrin Madhu Kumar and Erica Turner 2019 ldquoAmericans and Privacy Concerned Confused and Feeling Lack of Control Over Their Personal Informationrdquo Pew Research Center Internet Science amp Tech (blog) November 15 2019 httpswwwpewresearchorginternet20191115americans-and-privacy- concerned-confused-and-feeling-lack-of-control-over-their-personal-infor mation
Barth Susanne and Menno D T de Jong 2017 ldquoThe Privacy Paradox ndash Inves-tigating Discrepancies between Expressed Privacy Concerns and Actual Online Behavior ndash A Systematic Literature Reviewrdquo Telematics and Infor-matics 34 (7) 1038ndash58 httpsdoiorg101016jtele201704013
Bernstein Justin Taylor A Holroyd Jessica E Atwell Joseph Ali and Rupali J Limaye 2019 ldquoRockland Countyrsquos Proposed Ban against Unvaccinated Minors Balancing Disease Control Trust and Libertyrdquo Vaccine June httpsdoiorg101016jvaccine201905093
Berry Scott M Elizabeth A Petzold Peter Dull Nathan M Thielman Coleen K Cunningham G Ralph Corey Micah T McClain et al 2016 ldquoA Response Adaptive Randomization Platform Trial for Efficient Evalua-tion of Ebola Virus Treatments A Model for Pandemic Responserdquo Clinical Trials (London England) 13 (1) 22ndash30 httpsdoiorg101177 1740774515621721
Beukenhorst AL DM Schultz J McBeth R Lakshminarayana JC Sergeant and WG Dixon 2017 ldquoUsing Smartphones for Research Outside Clinical
Works Cited 129
Settings How Operating Systems App Evelopers and Users Determine Geolocation Data Quality in MHealth Studiesrdquo In MEDINFO 2017 Precision Healthcare through Infomatics 10ndash14 IOS Press httpebooks iospressnlvolumearticle48095
Bradshaw Tim 2020 ldquo2 Billion Phones Cannot Use Google and Apple Con-tact-Tracing Tech | Ars Technicardquo Ars Technica April 20 2020 httpsarstechnicacomtech-policy2020042-billion-phones-cannot-use-google-and-apple-contract-tracing-tech
Canca Cansu 2020 ldquoWhy lsquoMandatory Privacy-Preserving Digital Contact Tracingrsquo Is the Ethical Measure against COVID-19rdquo Medium May 4 2020 httpsmediumcomcansucancawhy-mandatory-privacy- preserving-digital-contact-tracing-is-the-ethical-measure-against- covid-19-a0d143b7c3b6
Cavalier Robert J ed 2011 Approaching Deliberative Democracy Theory and Practice Carnegie Mellon University Press
Cavoukian Ann 2010 ldquoPrivacy by Design The 7 Foundational Principlesrdquo Ontario Canada Information and Privacy Commissioner of Canada httpsiapporgmediapdfresource_centerpbd_implement_7found_ principlespdf
CDC 2005 ldquoVI Plan for Surveillance of Contacts of SARS Case Supplement B SARS Surveillancerdquo Centers for Disease Control and Prevention httpswwwcdcgovsarsguidanceb-surveillancecontactshtml
mdashmdashmdash 2020a ldquoContact Tracingrdquo Get and Keep America Open Supporting States Tribes Localities and Territories Centers for Disease Control and Prevention httpswwwcdcgovcoronavirus2019-ncovphpopen-americacontact-tracinghtml
mdashmdashmdash 2020b ldquoContact Tracing Part of a Multipronged Approach to Fight the COVID-19 Pandemicrdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention httpswwwcdcgovcoronavirus2019- ncovphpprinciples-contact-tracinghtml
mdashmdashmdash 2020c ldquoInterim Guidance for Businesses and Employersrdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention httpswwwcdcgovcoronavirus2019-ncovcommunityguidance-business- responsehtml
mdashmdashmdash 2020d ldquoOperational Considerations for the Identification of Healthcare Workers and Inpatients with Suspected COVID-19 in Non-US Healthcare Settingsrdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention httpswwwcdcgovcoronavirus2019-ncovhcpnon-us-settingsguidance-identify-hcw-patientshtml
130 Works Cited
mdashmdashmdash 2020e ldquoPreliminary Criteria for the Evaluation of Digital Contact Trac-ing Tools for COVID-19rdquo Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention httpswwwcdcgovcoronavirus 2019-ncovdownloadsphpprelim-eval-criteria-digital-contact-tracingpdf
mdashmdashmdash 2020f ldquoLegal Authorities for Isolation and Quarantinerdquo Centers for Disease Control and Prevention httpswwwcdcgovquarantine aboutlawsregulationsquarantineisolationhtml
mdashmdashmdash 2020g ldquoDiscontinuation of Isolation for Persons with COVID-19 Not in Healthcare Settingsrdquo Interim Guidance Coronavirus Disease 2019 (COVID-19) Centers for Disease Control and Prevention httpswww cdcgovcoronavirus2019-ncovhcpdisposition-in-home-patientshtml
mdashmdashmdash 2020h ldquoCOVID-19 Provisional CountsndashWeekly Updates by Select Demographic and Geographic Characteristicsrdquo CDC National Vital Statis-tics System httpswwwcdcgovnchsnvssvsrrcovid_weeklyindexhtm
Center for Survey Measurement (CSM) MEMORANDUM FOR Associate Directorate for Research and Methodology (ADRM) 2017 ldquoRespondent Confidentiality Concernsrdquo September 20 2017 httpswww2censusgovcacnacmeetings2017-11Memo-Regarding-Respondent-Confidentiality- Concernspdf
Cheng Hao-Yuan Shu-Wan Jian Ding-Ping Liu Ta-Chou Ng Wan-Ting Huang and Hsien-Ho Lin 2020 ldquoContact Tracing Assessment of COVID-19 Transmission Dynamics in Taiwan and Risk at Different Exposure Periods Before and After Symptom Onsetrdquo JAMA Internal Medicine May httpsdoiorg101001jamainternmed20202020
Cochrane Emily Claire Cain Miller and Jim Tankersley 2020 ldquoTrump Admin-istration Scales Back Paid Leave in Coronavirus Relief Lawrdquo The New York Times April 2 2020 sec US httpswwwnytimescom20200402uspoliticscoronavirus-paid-leavehtml
Crocker Andrew Kurt Opsahl and Bennett Cyphers 2020 ldquoThe Challenge of Proximity Apps For COVID-19 Contact Tracingrdquo Electronic Fron-tier Foundation April 10 2020 httpswwwefforgdeeplinks202004challenge-proximity-apps-covid-19-contact-tracing
Danquah Lisa O Nadia Hasham Matthew MacFarlane Fatu E Conteh Fatoma Momoh Andrew A Tedesco Amara Jambai David A Ross and Helen A Weiss 2019 ldquoUse of a Mobile Application for Ebola Contact Tracing and Monitoring in Northern Sierra Leone A Proof-of-Concept Studyrdquo BMC Infectious Diseases 19 (1) 810 httpsdoiorg101186s12879-019-4354-z
de Jong Bouke C Badou M Gaye Jeroen Luyten Bart van Buitenen Emman-
Works Cited 131
uel Andreacute Conor J Meehan Cian OrsquoSiochain et al 2019 ldquoEthical Considerations for Movement Mapping to Identify Disease Transmission Hotspotsrdquo Emerging Infectious Diseases 25 (7) httpsdoiorg103201eid2507181421
Devakumar Delan Geordan Shannon Sunil S Bhopal and Ibrahim Abu-bakar 2020 ldquoRacism and Discrimination in COVID-19 Responsesrdquo Lancet (London England) 395 (10231) 1194 httpsdoiorg101016S0140-6736(20)30792-3
Dixit Pranav 2020 ldquoIndiarsquos Contact Tracing App Is All But Mandatory So This Programmer Hacked It So That He Always Appears Saferdquo BuzzFeed News May 12 2020 httpswwwbuzzfeednewscomarticlepranavdixitindia-aarogya-setu-hacked
Doerr Megan Christine Suver and John Wilbanks 2016 ldquoDeveloping a Trans-parent Participant-Navigated Electronic Informed Consent for Mobile- Mediated Researchrdquo SSRN Scholarly Paper ID 2769129 Rochester NY Social Science Research Network httpsdoiorg102139ssrn2769129
Dredze Mark Michael J Paul Shane Bergsma and Hieu Tran 2013 ldquoCarmen A Twitter Geolocation System with Applications to Public Healthrdquo AAAI Workshops Workshops at the Twenty-Seventh AAAI Conference on Artifi-cial Intelligence httpswwwaaaiorgocsindexphpWSAAAIW13paperview70856497
Eckhoff Philip A and Andrew J Tatem 2015 ldquoDigital Methods in Epidemi-ology Can Transform Disease Controlrdquo International Health 7 (2) 77ndash78 httpsdoiorg101093inthealthihv013
EEOC 2020 ldquoWhat You Should Know About COVID-19 and the ADA the Rehabilitation Act and Other EEO Lawsrdquo US Equal Employment Opportunity Commission May 7 2020 httpswwweeocgovwyskwhat-you-should-know-about-covid-19-and-ada-rehabilitation-act-and-other-eeo-laws
Electronic Privacy Information Center Testimony to Congress 2020 ldquoEPIC to Congress Establish Privacy Safeguards for Digital Contact Tracingrdquo April 15 2020 httpsepicorgtestimonycongressEPIC-HEC-Contact-Tracing-Apr2020pdf
Elliott Douglas J Ana Kreacic Lorenzo Milans del Bosch and Lisa Quest 2020 ldquoData-Sharing in the Time of Coronavirusrdquo Oliver Wyman Forum Accessed May 19 2020 httpswwwoliverwymanforumcomfuture-of-data2020aprdata-sharing-in-the-time-of-coronavirushtml
Faden Ruth R and Tom L Beauchamp 1986 A History and Theory of Informed Consent Oxford University Press
132 Works Cited
FCC and FTC 2017 ldquoFCC-FTC Consumer Protection Memorandum of Understandingrdquo Federal Communications Commission and Federal Trade Commission httpswwwftcgovsystemfilesdocumentscooperation_agreements151116ftcfcc-moupdf
Ferretti Luca Chris Wymant Michelle Kendall Lele Zhao Anel Nurtay Lucie Abeler-Doumlrner Michael Parker David Bonsall and Christophe Fraser 2020 ldquoQuantifying SARS-CoV-2 Transmission Suggests Epidemic Control with Digital Contact Tracingrdquo Science 368 (6491) httpsdoiorg101126scienceabb6936
Fishkin James S and Peter Laslett eds 2003 Debating Deliberative Democ-racy Blackwell Publisher Ltd httpsonlinelibrarywileycomdoibook 1010029780470690734
Flanagan Mary Daniel C Howe and Helen Nissenbaum 2008 ldquoEmbodying Values in Technology Theory and Practicerdquo In Information Technology and Moral Philosophy edited by Jeroen van den Hoven and John Weckert 322ndash53 Cambridge University Press
Fraccaro Paolo Anna Beukenhorst Matthew Sperrin Simon Harper Jasper Palmier-Claus Shocircn Lewis Sabine N Van der Veer and Niels Peek 2019 ldquoDigital Biomarkers from Geolocation Data in Bipolar Disorder and Schizophrenia A Systematic Reviewrdquo Journal of the American Medical Informatics Association 26 (11) 1412ndash20 httpsdoiorg101093jamiaocz043
Fraser Christophe Lucie Abeler-Doumlrner Luca Ferretti Michael Parker Michelle Kendall and David Bonsall 2020 ldquoDigital Contact Tracing Comparing the Capabilities of Centralised and Decentralised Data Architectures to Effectively Suppress the COVID-19 Epidemic Whilst Maximising Freedom of Movement and Maintaining Privacyrdquo httpsgithubcomBDI-pathogenscovid-19_instant_tracingblobmasterCentralised20and20decen-tralised20systems20for20contact20tracingpdf
FTC 2012 ldquoProtecting Consumer Privacy in an Era of Rapid Changerdquo FTC Report Recommendations for Businesses and Policymakers Federal Trade Commission httpswwwftcgovsitesdefaultfilesdocumentsreports federal-trade-commission-report-protecting-consumer-privacy- era-rapid-change-recommendations120326privacyreportpdf
mdashmdashmdash 2020 ldquoPrivacy amp Data Security Update 2019rdquo Federal Trade Commis-sion httpswwwftcgovsystemfilesdocumentsreportsprivacy-data- security-update-20192019-privacy-data-security-report-508pdf
Furlanello Cesare Stefano Merler Stefano Menegon Sebastiano Mancuso and Gianni Bertiato 2002 ldquoNew WEBGIS Technologies for Geo-Location of
Works Cited 133
Epidemiological Data An Application for the Surveillance of the Risk of Lyme Borreliosis Diseaserdquo Giornale Italiano Di Aritmologia e Cardiosti-molazione Special Issue Proceedings of the 15th Int Congress ldquoThe New Frontiers of Arrhythmiasrdquo 5 (1) 241ndash45
Gan Nectar and David Culver 2020 ldquoChina Is Fighting the Coronavirus with a Digital QR Code Herersquos How It Worksrdquo CNN Business April 16 2020 httpswwwcnncom20200415asiachina-coronavirus-qr-code-intl-hnkindexhtml
Gibson Dustin G Adaeze C Wosu George William Pariyo Saifuddin Ahmed Joseph Ali Alain B Labrique Iqbal Ansary Khan Elizeus Rutebemberwa Meerjady Sabrina Flora and Adnan A Hyder 2019 ldquoEffect of Airtime Incentives on Response and Cooperation Rates in Non-Communicable Disease Interactive Voice Response Surveys Randomised Controlled Trials in Bangladesh and Ugandardquo BMJ Global Health 4 (5) e001604 httpsdoiorg101136bmjgh-2019-001604
Giglio Mike 2020 ldquoWould You Sacrifice Your Privacy to Get Out of Quarantinerdquo The Atlantic April 22 2020 httpswwwtheatlanticcompoliticsarchive202004coronavirus-pandemic-privacy-civil-liberties-911 609172
Guariglia Matthew 2020 ldquoThe Dangers of COVID-19 Surveillance Proposals to the Future of Protestrdquo Electronic Frontier Foundation April 29 2020 httpswwwefforgdeeplinks202004some-covid-19-surveillance-propos als-could-harm-free-speech-after-covid-19
Hadavas Chloe 2020 ldquoHow Effective Are Contact Tracing Appsrdquo Slate Maga-zine May 13 2020 httpsslatecomtechnology202005contact- tracing-apps-less-effective-icelandhtml
Hamilton Isobel Asher 2020 ldquoPoland Made an App That Forces Coronavirus Patients to Take Regular Selfies to Prove Theyrsquore Indoors or Face a Police Visitrdquo Business Insider March 23 2020 httpswwwbusinessinsidercompoland-app-coronavirus-patients-mandaotory-selfie-2020-3
Hargittai Eszter Minh Hao Nguyen Jaelle Fuchs Jonathan Gruber Will Marler Amanda Hunsaker and Goumlkccedile Karaoglu 2020 ldquoCovid-19 Study on Digital Media and the Coronavirus Pandemicrdquo Internet Use and Society Division Institute of Communication and Media Research University of Zurich httpwebuseorgcovid
Hargittai Eszter and Elissa Redmiles 2020 ldquoWill Americans Be Willing to Install COVID-19 Tracking Appsrdquo Scientific American Blog Net-work April 28 2020 httpsblogsscientificamericancomobservationswill-americans-be-willing-to-install-covid-19-tracking-apps
134 Works Cited
Hart Vi Divya Siddarth Bethan Cantrell Lila Tretikov Peter Eckersley John Langford Scott Leibrand et al 2020 ldquoOutpacing the Virus Dig-ital Response to Containing the Spread of COVID-19 While Mitigat-ing Privacy Risksrdquo Whitepaper 5 COVID-19 Rapid Response Impact Initiative Edmond J Safra Center for Ethics httpsdrivegooglecomfiled1vIN2AX-DDNW-S0aHq8xs0RJ2jkR_CckXview
Hemming K T P Haines P J Chilton A J Girling and R J Lilford 2015 ldquoThe Stepped Wedge Cluster Randomised Trial Rationale Design Anal-ysis and Reportingrdquo BMJ 350 (February) httpsdoiorg101136bmjh391
Hendrix Steve and Ruth Eglash 2020 ldquoIsrael Is Using Cellphone Surveil-lance to Warn Citizens You May Already Be Infectedrdquo Washington Post Accessed May 19 2020 httpswwwwashingtonpostcomworldmiddle_eastisrael-is-using-cellphone-surveillance-to-warn-citizens-you-may- already-be-infected2020031968267294-69e7-11ea-b199-3a9799c54 512_storyhtml
Heneghan Carl Jon Brassey and Tom Jefferson 2020 ldquoCOVID-19 What Proportion Are Asymptomaticrdquo Centre for Evidence-Based Medi-cine httpswwwcebmnetcovid-19covid-19-what-proportion-are- asymptomatic
HHS 2019 ldquoPublic Health Emergency Declarationrdquo Public Health Emergency Accessed May 19 2020 httpswwwphegovPreparednesslegalPagesphedeclarationaspx
Hickey Matt 2014 ldquoCarriers Can Now Install Apps On Android Handsets Without Customersrsquo Permissionrdquo Forbes December 1 2014 httpswww forbescomsitesmatthickey20141201carriers-can-now-install- apps-on-android-handsets-without-customers-permission
Hinch Robert Will Probert Anel Nurtay Michelle Kendall Chris Wymant Matthew Hall Katrina Lythgoe et al 2020 ldquoEffective Configurations of a Digital Contact Tracing App A Report to NHSXrdquo fileUsersameliahoodDownloadsReport20-20Effective20App20Configurations20(1)pdf
Ingram David 2020 ldquoApple Google Push Makers of Coronavirus Apps Not to Record User Locationrdquo May 4 2020 NBC News Accessed May 19 2020 httpswwwnbcnewscomtechtech-newscoronavirus-apps-won-t-be-able- record-users-location-apple-n1199586
Iwaya Leonardo H Jane Li Simone Fischer-Huumlbner Rose-Mharie Aringhlfeldt and Leonardo A Martucci 2019 ldquoE-Consent for Data Privacy Consent Management for Mobile Health Technologies in Public Health Surveys and
Works Cited 135
Disease Surveillancerdquo Studies in Health Technology and Informatics 264 (August) 1223ndash27 httpsdoiorg103233SHTI190421
ldquoJoint Statement on Contact Tracingrdquo 2020 April 19 httpscryptobriefingcomwp-contentuploads202004Joint-Statement-from-Researcherspdf
Kahn Gilmor Daniel 2020 ldquoPrinciples for Technology-Assisted Contact- Tracingrdquo White Paper American Civil Liberties Union httpswwwaclu orgreportaclu-white-paper-principles-technology-assisted-contact-tracing
Kim Max S 2020 ldquoSeoulrsquos Radical Experiment in Digital Contact Tracingrdquo The New Yorker April 17 2020 httpswwwnewyorkercomnewsnews-deskseouls-radical-experiment-in-digital-contact-tracing
Kim Nemo 2020 ldquolsquoMore Scary than Coronavirusrsquo South Korearsquos Health Alerts Expose Private Livesrdquo The Guardian March 6 2020 sec World news httpswwwtheguardiancomworld2020mar06more-scary- than-coronavirus-south-koreas-health-alerts-expose-private-lives
Kirzinger Ashley Liz Hamel Cailey Muntildeana Audrey Kearney and Mollyann Brodie 2020 ldquoKFF Health Tracking Poll ndash Late April 2020 Coronavirus Social Distancing and Contact Tracingrdquo Kaiser Family Foundation (blog) April 24 2020 httpswwwkfforgcoronavirus-covid-19issue-briefkff-health-tracking-poll-late-april-2020
Knobel Cory and Geoffrey C Bowker 2011 ldquoComputing Ethics Value in Designrdquo Communications of the ACM 54 (7) 26ndash28 httpsdoiorgdoi10114519657241965735
Lee Ramon Rosa R Cui Kathryn E Muessig Harsha Thirumurthy and Joseph D Tucker 2014 ldquoIncentivizing HIVSTI Testing A Systematic Review of the Literaturerdquo AIDS and Behavior 18 (5) 905ndash12 https doiorg101007s10461-013-0588-8
Leprince-Ringuet Daphne 2020 ldquoContact-Tracing Apps Why the NHS Said No to Apple and Googlersquos Planrdquo ZDNet April 28 2020 httpswwwzdnetcomarticlecontact-tracing-apps-why-the-nhs-said-no-to-apple-and- googles-plan
Lovejoy Ben 2020 ldquoNHS Contact Tracing App Hands-on as the Test Goes Liverdquo 9to5Mac (blog) May 7 2020 https9to5maccom20200507nhs-contact-tracing
Mahmood Sultan Khaled Hasan Michelle Colder Carras and Alain Labri-que 2020 ldquoGlobal Preparedness Against COVID-19 We Must Leverage the Power of Digital Healthrdquo JMIR Public Health and Surveillance 6 (2) e18980 httpsdoiorg10219618980
Mathews Simon C Michael J McShea Casey L Hanley Alan Ravitz Alain B Labrique and Adam B Cohen 2019 ldquoDigital Health A Path to
136 Works Cited
Validationrdquo Npj Digital Medicine 2 (1) 1ndash9 httpsdoiorg101038s41746-019-0111-3
Mello By Michelle M and C Jason Wang 2020 ldquoEthics and Governance for Digital Disease Surveillancerdquo Science May httpsdoiorg101126science abb9045
Mills Rodrigo Chris 2020 ldquoDigital Contact Tracing Is Becoming Available but Is It Effectiverdquo The Hill May 7 2020 httpsthehillcompolicytechnolo-gy496498-digital-contact-tracing-is-becoming-available-but-is-it-effective
MIT nd ldquoPrivate Kit Safe Paths Privacy-by-Design Covid19 Solutions Using GPS+Bluetooth for Citizens and Public Health Officialsrdquo Safepaths Accessed May 20 2020 httpssafepathsmitedu
ldquoMobile Cellular Subscriptions (per 100 People)rdquo 2018 World Telecommuni-cationICT Development Report International Telecommunication Union httpsdataworldbankorgindicatorITCELSETSP2
Moore Sarah Anne-Marie Tasseacute Adrian Thorogood Ingrid Winship Marsquon Zawati and Megan Doerr 2017 ldquoConsent Processes for Mobile App Mediated Research Systematic Reviewrdquo JMIR MHealth and UHealth 5 (8) e126 httpsdoiorg102196mhealth7014
Morse Jack 2020 ldquoNorth Dakota Launched a Contact-Tracing App Itrsquos Not Going Wellrdquo Mashable Accessed May 19 2020 httpsmashablecomarticlenorth-dakota-contact-tracing-app
Muller Robert T 2020 ldquoCOVID-19 Brings a Pandemic of Conspiracy Theo-riesrdquo Psychology Today April 24 2020 httpswwwpsychologytodaycomblogtalking-about-trauma202004covid-19-brings-pandemic-conspiracy- theories
Mulligan Stephen P and Chris D Linebaugh 2019 ldquoData Protection Law An Overviewrdquo R45631 Congressional Research Service httpscrsreports congressgovproductpdfRR45631
National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research 1979 ldquoThe Belmont Report Ethical Principles and Guidelines for the Protection of Human Subjects of Researchrdquo US Dept of Health amp Human Services Office of Human Research Protections httpswwwhhsgovohrpregulations-and-policybelmont-reportread-the-belmont-reportindexhtml
NDDoH 2020 ldquoNorth Dakota Launches Care19 App to Combat COVID-19rdquo North Dakota Department of Health April 7 2020 httpswwwhealth ndgovnewsnorth-dakota-launches-care19-app-combat-covid-19
NYC DOH 2020 ldquoAge-Adjusted Rates of Lab Confirmed COVID-19 Non-hospitalized Cases Estimated Non-Fatal Hospitalized Cases and Persons Known to Have Died per 100000 by RaceEthnicity Grouprdquo New York
Works Cited 137
City Department of Health httpswww1nycgovassetsdohdownloadspdfimmcovid-19-deaths-race-ethnicity-04242020-1pdf
OrsquoRielly Michael 2018 ldquoFCC Regulatory Free Arenardquo Federal Communica-tions Commission June 1 2018 httpswwwfccgovnews-eventsblog 20180601fcc-regulatory-free-arena
OSHA 2020 ldquoGuidance on Preparing Workplaces for COVID-19rdquo OSHA 3990-03 2020 Department of Labor Occupational Safety and Health Act of 1970 httpswwwoshagovPublicationsOSHA3990pdf
Pallmann Philip Alun W Bedding Babak Choodari-Oskooei Munyaradzi Dimairo Laura Flight Lisa V Hampson Jane Holmes et al 2018 ldquoAdaptive Designs in Clinical Trials Why Use Them and How to Run and Report Themrdquo BMC Medicine 16 (1) 29 httpsdoiorg101186s12916-018-1017-7
Pew Research Center 2017 ldquoUS Muslims Concerned About Their Place in Society but Continue to Believe in the American Dreamrdquo Religion amp Public Life Pew Research Center httpswwwpewforumorg20170726findings-from-pew-research-centers-2017-survey-of-us-muslims
mdashmdashmdash 2020 ldquoDemographics of Mobile Device Ownership and Adoption in the United Statesrdquo Internet amp Technology Pew Research Center Accessed May 19 2020 httpswwwpewresearchorginternetfact-sheetmobile
PHLS 2002 ldquoPrinciples of the Ethical Practice of Public Healthrdquo Public Health Leadership Society httpswwwaphaorg-mediafilespdfmembergroupsethicsethics_brochureashx
PIH 2020a ldquoPart I Testing Contact Tracing and Community Management of COVID-19rdquo PIH Guide | COVID-19 STOP COVID Partners in Health httpswwwpihorgsitesdefaultfiles2020-04PIH_Guide_COVID_Part_I_Testing_Tracing_Community_Managment_4_4pdf
mdashmdashmdash 2020b ldquoBox It Inrdquo PIH Guide | COVID-19 Partners in Health httpspreventepidemicsorgcovid19resourcesbox-it-in
Resolve to Save Lives 2020 ldquoCOVID-19 Contact Tracing Playbookrdquo Vital Strategies httpscontacttracingplaybookresolvetosavelivesorg
Reston Maeve Kristina Sgueglia and Cheri Mossburg 2020 ldquoGovernors on East and West Coasts Form Pacts to Decide When to Reopen Economiesrdquo CNN Politics April 13 2020 httpswwwcnncom20200413politicsstates-band-together-reopening-plansindexhtml
Rithalia Amber Catriona McDaid Sara Suekarran Lindsey Myers and Amanda Sowden 2009 ldquoImpact of Presumed Consent for Organ Donation on Donation Rates A Systematic Reviewrdquo BMJ 338 (January) httpsdoiorg101136bmja3162
Rodrigues Rafaela Alina Husain Amanda Couture-Carron Leslye E Orloff
138 Works Cited
and Nawal H Ammar 2018 ldquoPromoting Access to Justice for Immigrant and Limited English Proficient Crime Victims in an Age of Increased Immi-gration Enforcement Initial Report from a 2017 National Surveyrdquo Wash-ington DC National Immigrant Womenrsquos Advocacy Project American University Washington College of Law httpniwaplibrarywclamericaneduwp-contentuploadsImmigrant-Access-to-Justice-National-Reportpdf
Simpson Erin and Adam Conner 2020 ldquoDigital Contact Tracing to Contain the Coronavirusrdquo Center for American Progress httpswww americanprogressorgissuestechnology-policynews20200422483521digital-contact-tracing-contain-coronavirus
Singer Eleanor and Cong Ye 2013 ldquoThe Use and Effects of Incentives in Surveysrdquo The ANNALS of the American Academy of Political and Social Science 645 (1) 112ndash41 httpsdoiorg1011770002716212458082
Thornton Rebecca L 2008 ldquoThe Demand for and Impact of Learning HIV Statusrdquo The American Economic Review 98 (5) 1829ndash63 httpsdoi org101257aer9851829
US DOE 2009 ldquoState Regulation of Private Schoolsrdquo US Dept of Education Office of Non-Public Education httpswww2edgovadminscommchoiceregprivschlregprivschlpdf
Valentino-DeVries Jennifer 2020 ldquoCellphone Carriers Face $200 Million Fine for Not Protecting Location Datardquo The New York Times February 28 2020 sec Technology httpswwwnytimescom20200228technologyfcc-cellphones-location-data-fineshtml
Valentino-DeVries Jennifer Natasha Singer and Aaron Krolik 2020 ldquoA Scramble for Virus Apps That Do No Harmrdquo The New York Times April 29 2020 sec Business httpswwwnytimescom20200429businesscoronavirus-cellphone-apps-contact-tracinghtml
Washington PostndashUMD ldquoWashington Post-University of Maryland National Pollrdquo 2020 Washington Post April 21 2020 httpswwwwashingtonpostcomcontextwashington-post-university-of-maryland-national-poll-april-21-26-20203583b4e9-66be-4ed6-a457-f6630a550ddf
Watson Crystal Anita Cicero James Blumenstock and Michael Fraser 2020 ldquoA National Plan to Enable Comprehensive COVID-19 Case Finding and Contact Tracing in the USrdquo Johns Hopkins Bloomberg School of Public Health Center for Health Security and the Association of State and Terri-torial Health Officials httpswwwcenterforhealthsecurityorgour-workpubs_archivepubs-pdfs2020200410-national-plan-to-contact-tracingpdf
Works Cited 139
WHO 2017 ldquoWHO Guidelines on Ethical Issues in Public Health Surveillancerdquo World Health Organization httpswwwwhointethicspublicationspublic-health-surveillanceen
mdashmdashmdash 2020 ldquoCoronavirus Disease 2019 (COVID-19)rdquo Situation Report 73 World Health Organization httpswwwwhointdocsdefault-sourcecoro-navirusesituation-reports20200402-sitrep-73-covid-19pdf
- _iy90yrtqgtxz
- _GoBack
- Acknowledgments
- Preface
- Lead Authors and Contributors
- Acronyms and Abbreviations
- Summary
-
- Introduction
- DCTT Features Functions and Potential Applications
- Summary of Recommendations
- Summary of Analysis
-
- Introduction
-
- Guiding Principles for the Use of Digital Public Health Technologies for Pandemic Response
-
- Public Health Perspective
-
- Types of Information Collected through Contact Tracing
- How Contact Tracing Information Informs Public Health Action
- Characteristics That Make Data Useful to Public Health for Reducing Disease Transmission
-
- Digital Technology and Contact Tracing
-
- Characteristics of SARS-CoV-2 Relevant to Candidate Digital Solutions
- Previously Existing Contact Tracing Technologies
- Introduction of Novel Digital Contact Tracing Technologies
- Relevant Differences between Manual and Digital Contact Tracing
-
- Ethics of Designing and Using DCTT
-
- Justifying the Use of DCTT Systems
- Monitoring and Evaluating Technologies to Inform Policy and Practice
- Public Trust and Public Attitudes
- Designing Flexible Technology to Maximize Public Health Utility While Respecting Other Values
- Policy Positions to Advance Widespread Use of Digital Contact Tracing Technologies
- Disclosure and AuthorizationConsent
- Promoting Equity and Fairness
- Instituting Transparent Governance and Oversight
-
- Legal Considerations
-
- Data Privacy and Data Security Laws
- Health Information Privacy
- Labor and Employment Privacy Rights
- Constitutional Privacy Rights
- Consent
- Anti-discrimination and Individual Freedom Laws
-
- Recommendations
-
- Public Health
- Ethics
- Legislative
-
- Resources
-
- US Government Response
- Other Governmental and Nongovernmental Organizations
- Digital Contact Tracing Experiences from Other Countries
- Specific Digital ProductsApps
- Polling
- Popular Press
- Commentaries
- Academic Literature
-
- Works Cited
-